Enter prerelease mode and version packages

fix(config): restores sonnet 3.7 as default main role.

.changeset/beige-rats-accept.md

@@ -0,0 +1,5 @@
+---
+'task-master-ai': patch
+---
+
+- Add support for Google Gemini models via Vercel AI SDK integration.

.changeset/blue-spies-kick.md

@@ -0,0 +1,5 @@
+---
+'task-master-ai': patch
+---
+
+Add xAI provider and Grok models support

.changeset/brave-seas-pull.md

@@ -1,6 +0,0 @@
----
-'task-master-ai': patch
----
-
-- Fixes shebang issue not allowing task-master to run on certain windows operating systems
-- Resolves #241 #211 #184 #193

.changeset/chubby-moose-stay.md

@@ -1,5 +0,0 @@
----
-'task-master-ai': patch
----
-
-Updates the parameter descriptions for update, update-task and update-subtask to ensure the MCP server correctly reaches for the right update command based on what is being updated -- all tasks, one task, or a subtask.

.changeset/cuddly-zebras-matter.md

@@ -0,0 +1,8 @@
+---
+'task-master-ai': minor
+---
+
+feat(expand): Enhance `expand` and `expand-all` commands
+
+- Integrate `task-complexity-report.json` to automatically determine the number of subtasks and use tailored prompts for expansion based on prior analysis. You no longer need to try copy-pasting the recommended prompt. If it exists, it will use it for you. You can just run `task-master update --id=[id of task] --research` and it will use that prompt automatically. No extra prompt needed. 
+- Change default behavior to *append* new subtasks to existing ones. Use the `--force` flag to clear existing subtasks before expanding. This is helpful if you need to add more subtasks to a task but you want to do it by the batch from a given prompt. Use force if you want to start fresh with a task's subtasks.

.changeset/curvy-candies-eat.md

@@ -0,0 +1,9 @@
+---
+'task-master-ai': patch
+---
+
+Better support for file paths on Windows, Linux & WSL.
+
+- Standardizes handling of different path formats (URI encoded, Windows, Linux, WSL).
+- Ensures tools receive a clean, absolute path suitable for the server OS.
+- Simplifies tool implementation by centralizing normalization logic.

.changeset/easy-toys-wash.md

@@ -0,0 +1,7 @@
+---
+'task-master-ai': minor
+---
+
+Adds support for the OpenRouter AI provider. Users can now configure models available through OpenRouter (requiring an `OPENROUTER_API_KEY`) via the `task-master models` command, granting access to a wide range of additional LLMs.
+    - IMPORTANT FYI ABOUT OPENROUTER: Taskmaster relies on AI SDK, which itself relies on tool use. It looks like **free** models sometimes do not include tool use. For example, Gemini 2.5 pro (free) failed via OpenRouter (no tool use) but worked fine on the paid version of the model. Custom model support for Open Router is considered experimental and likely will not be further improved for some time.
+

.changeset/every-stars-sell.md

@@ -0,0 +1,5 @@
+---
+'task-master-ai': patch
+---
+
+Add integration for Roo Code

.changeset/fine-monkeys-eat.md

@@ -0,0 +1,8 @@
+---
+'task-master-ai': patch
+---
+
+Improved update-subtask
+    - Now it has context about the parent task details
+    - It also has context about the subtask before it and the subtask after it (if they exist)
+    - Not passing all subtasks to stay token efficient

.changeset/fine-signs-add.md

@@ -0,0 +1,13 @@
+---
+'task-master-ai': patch
+---
+
+Improve and adjust `init` command for robustness and updated dependencies.
+
+- **Update Initialization Dependencies:** Ensure newly initialized projects (`task-master init`) include all required AI SDK dependencies (`@ai-sdk/*`, `ai`, provider wrappers) in their `package.json` for out-of-the-box AI feature compatibility. Remove unnecessary dependencies (e.g., `uuid`) from the init template.
+- **Silence `npm install` during `init`:** Prevent `npm install` output from interfering with non-interactive/MCP initialization by suppressing its stdio in silent mode.
+- **Improve Conditional Model Setup:** Reliably skip interactive `models --setup` during non-interactive `init` runs (e.g., `init -y` or MCP) by checking `isSilentMode()` instead of passing flags.
+- **Refactor `init.js`:** Remove internal `isInteractive` flag logic.
+- **Update `init` Instructions:** Tweak the "Getting Started" text displayed after `init`.
+- **Fix MCP Server Launch:** Update `.cursor/mcp.json` template to use `node ./mcp-server/server.js` instead of `npx task-master-mcp`.
+- **Update Default Model:** Change the default main model in the `.taskmasterconfig` template.

.changeset/gentle-views-jump.md

@@ -0,0 +1,5 @@
+---
+'task-master-ai': patch
+---
+
+Fixes an issue with add-task which did not use the manually defined properties and still needlessly hit the AI endpoint.

.changeset/mighty-mirrors-watch.md

@@ -0,0 +1,5 @@
+---
+'task-master-ai': minor
+---
+
+Adds model management and new configuration file .taskmasterconfig which houses the models used for main, research and fallback. Adds models command and setter flags. Adds a --setup flag with an interactive setup. We should be calling this during init. Shows a table of active and available models when models is called without flags. Includes SWE scores and token costs, which are manually entered into the supported_models.json, the new place where models are defined for support. Config-manager.js is the core module responsible for managing the new config."

.changeset/moody-pugs-grab.md

@@ -1,6 +0,0 @@
----
-'task-master-ai': patch
----
-
-- Fix `task-master init` polluting codebase with new packages inside `package.json` and modifying project `README`
-  - Now only initializes with cursor rules, windsurf rules, mcp.json, scripts/example_prd.txt, .gitignore modifications, and `README-task-master.md`

.changeset/neat-donkeys-shave.md

@@ -0,0 +1,5 @@
+---
+'task-master-ai': patch
+---
+
+Fixes an issue that prevented remove-subtask with comma separated tasks/subtasks from being deleted (only the first ID was being deleted). Closes #140

.changeset/nine-rocks-sink.md

@@ -0,0 +1,10 @@
+---
+'task-master-ai': patch
+---
+
+Improves next command to be subtask-aware
+    - The logic for determining the "next task" (findNextTask function, used by task-master next and the next_task MCP tool) has been significantly improved. Previously, it only considered top-level tasks, making its recommendation less useful when a parent task containing subtasks was already marked 'in-progress'.
+    - The updated logic now prioritizes finding the next available subtask within any 'in-progress' parent task, considering subtask dependencies and priority. 
+    - If no suitable subtask is found within active parent tasks, it falls back to recommending the next eligible top-level task based on the original criteria (status, dependencies, priority). 
+    
+This change makes the next command much more relevant and helpful during the implementation phase of complex tasks.

.changeset/ninety-ghosts-relax.md

@@ -0,0 +1,11 @@
+---
+'task-master-ai': minor
+---
+
+Adds custom model ID support for Ollama and OpenRouter providers.
+  - Adds the `--ollama` and `--openrouter` flags to `task-master models --set-<role>` command to set models for those providers outside of the support models list.
+  - Updated `task-master models --setup` interactive mode with options to explicitly enter custom Ollama or OpenRouter model IDs.
+  - Implemented live validation against OpenRouter API (`/api/v1/models`) when setting a custom OpenRouter model ID (via flag or setup).
+  - Refined logic to prioritize explicit provider flags/choices over internal model list lookups in case of ID conflicts.
+  - Added warnings when setting custom/unvalidated models.
+  - We obviously don't recommend going with a custom, unproven model. If you do and find performance is good, please let us know so we can add it to the list of supported models.

.changeset/ninety-wombats-pull.md

@@ -0,0 +1,5 @@
+---
+'task-master-ai': patch
+---
+
+Add `--status` flag to `show` command to filter displayed subtasks.

.changeset/polite-candles-follow.md

@@ -1,5 +0,0 @@
----
-'task-master-ai': minor
----
-
-Add `npx task-master-ai` that runs mcp instead of using `task-master-mcp``

.changeset/pre.json

@@ -0,0 +1,28 @@
+{
+  "mode": "pre",
+  "tag": "rc",
+  "initialVersions": {
+    "task-master-ai": "0.12.1"
+  },
+  "changesets": [
+    "beige-rats-accept",
+    "blue-spies-kick",
+    "cuddly-zebras-matter",
+    "curvy-candies-eat",
+    "easy-toys-wash",
+    "every-stars-sell",
+    "fine-monkeys-eat",
+    "fine-signs-add",
+    "gentle-views-jump",
+    "mighty-mirrors-watch",
+    "neat-donkeys-shave",
+    "nine-rocks-sink",
+    "ninety-ghosts-relax",
+    "ninety-wombats-pull",
+    "public-cooks-fetch",
+    "red-oranges-attend",
+    "tricky-papayas-hang",
+    "violet-papayas-see",
+    "violet-parrots-march"
+  ]
+}

.changeset/public-cooks-fetch.md

@@ -0,0 +1,7 @@
+---
+'task-master-ai': minor
+---
+
+Integrate OpenAI as a new AI provider.
+    - Enhance `models` command/tool to display API key status.
+    - Implement model-specific `maxTokens` override based on `supported-models.json` to save you if you use an incorrect max token value.

.changeset/red-oranges-attend.md

@@ -0,0 +1,5 @@
+---
+'task-master-ai': patch
+---
+
+Fix ERR_MODULE_NOT_FOUND when trying to run MCP Server

.changeset/tricky-papayas-hang.md

@@ -0,0 +1,9 @@
+---
+'task-master-ai': minor
+---
+Tweaks Perplexity AI calls for research mode to max out input tokens and get day-fresh information
+    - Forces temp at 0.1 for highly deterministic output, no variations
+    - Adds a system prompt to further improve the output
+    - Correctly uses the maximum input tokens (8,719, used 8,700) for perplexity
+    - Specificies to use a high degree of research across the web
+    - Specifies to use information that is as fresh as today; this support stuff like capturing brand new announcements like new GPT models and being able to query for those in research. 🔥

.changeset/violet-papayas-see.md

@@ -0,0 +1,5 @@
+---
+'task-master-ai': patch
+---
+
+Fix --task to --num-tasks in ui + related tests - issue #324

.changeset/violet-parrots-march.md

@@ -0,0 +1,9 @@
+---
+'task-master-ai': patch
+---
+
+Adds a 'models' CLI and MCP command to get the current model configuration, available models, and gives the ability to set main/research/fallback models." 
+    - In the CLI, `task-master models` shows the current models config. Using the `--setup` flag launches an interactive set up that allows you to easily select the models you want to use for each of the three roles. Use `q` during the interactive setup to cancel the setup.
+    - In the MCP, responses are simplified in RESTful format (instead of the full CLI output). The agent can use the `models` tool with different arguments, including `listAvailableModels` to get available models. Run without arguments, it will return the current configuration. Arguments are available to set the model for each of the three roles. This allows you to manage Taskmaster AI providers and models directly from either the CLI or MCP or both.
+    - Updated the CLI help menu when you run `task-master` to include missing commands and .taskmasterconfig information.
+    - Adds `--research` flag to `add-task` so you can hit up Perplexity right from the add-task flow, rather than having to add a task and then update it.

.cursor/mcp.json

@@ -1,17 +1,18 @@
 {
 	"mcpServers": {
-		"taskmaster-ai": {
+		"task-master-ai": {
 			"command": "node",
 			"args": ["./mcp-server/server.js"],
 			"env": {
-				"ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
-				"PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY_HERE",
-				"MODEL": "claude-3-7-sonnet-20250219",
-				"PERPLEXITY_MODEL": "sonar-pro",
-				"MAX_TOKENS": 64000,
-				"TEMPERATURE": 0.2,
-				"DEFAULT_SUBTASKS": 5,
-				"DEFAULT_PRIORITY": "medium"
+				"ANTHROPIC_API_KEY": "ANTHROPIC_API_KEY_HERE",
+				"PERPLEXITY_API_KEY": "PERPLEXITY_API_KEY_HERE",
+				"OPENAI_API_KEY": "OPENAI_API_KEY_HERE",
+				"GOOGLE_API_KEY": "GOOGLE_API_KEY_HERE",
+				"XAI_API_KEY": "XAI_API_KEY_HERE",
+				"OPENROUTER_API_KEY": "OPENROUTER_API_KEY_HERE",
+				"MISTRAL_API_KEY": "MISTRAL_API_KEY_HERE",
+				"AZURE_OPENAI_API_KEY": "AZURE_OPENAI_API_KEY_HERE",
+				"OLLAMA_API_KEY": "OLLAMA_API_KEY_HERE"
 			}
 		}
 	}

.cursor/rules/ai_providers.mdc

@@ -0,0 +1,155 @@
+---
+description: Guidelines for managing Task Master AI providers and models.
+globs: 
+alwaysApply: false
+---
+# Task Master AI Provider Management
+
+This rule guides AI assistants on how to view, configure, and interact with the different AI providers and models supported by Task Master. For internal implementation details of the service layer, see [`ai_services.mdc`](mdc:.cursor/rules/ai_services.mdc).
+
+-   **Primary Interaction:**
+    -   Use the `models` MCP tool or the `task-master models` CLI command to manage AI configurations. See [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc) for detailed command/tool usage.
+
+-   **Configuration Roles:**
+    -   Task Master uses three roles for AI models:
+        -   `main`: Primary model for general tasks (generation, updates).
+        -   `research`: Model used when the `--research` flag or `research: true` parameter is used (typically models with web access or specialized knowledge).
+        -   `fallback`: Model used if the primary (`main`) model fails.
+    -   Each role is configured with a specific `provider:modelId` pair (e.g., `openai:gpt-4o`).
+
+-   **Viewing Configuration & Available Models:**
+    -   To see the current model assignments for each role and list all models available for assignment:
+        -   **MCP Tool:** `models` (call with no arguments or `listAvailableModels: true`)
+        -   **CLI Command:** `task-master models`
+    -   The output will show currently assigned models and a list of others, prefixed with their provider (e.g., `google:gemini-2.5-pro-exp-03-25`).
+
+-   **Setting Models for Roles:**
+    -   To assign a model to a role:
+        -   **MCP Tool:** `models` with `setMain`, `setResearch`, or `setFallback` parameters.
+        -   **CLI Command:** `task-master models` with `--set-main`, `--set-research`, or `--set-fallback` flags.
+    -   **Crucially:** When providing the model ID to *set*, **DO NOT include the `provider:` prefix**. Use only the model ID itself.
+        -   ✅ **DO:** `models(setMain='gpt-4o')` or `task-master models --set-main=gpt-4o`
+        -   ❌ **DON'T:** `models(setMain='openai:gpt-4o')` or `task-master models --set-main=openai:gpt-4o`
+    -   The tool/command will automatically determine the provider based on the model ID.
+
+-   **Setting Custom Models (Ollama/OpenRouter):**
+    -   To set a model ID not in the internal list for Ollama or OpenRouter:
+        -   **MCP Tool:** Use `models` with `set<Role>` and **also** `ollama: true` or `openrouter: true`.
+            -   Example: `models(setMain='my-custom-ollama-model', ollama=true)`
+            -   Example: `models(setMain='some-openrouter-model', openrouter=true)`
+        -   **CLI Command:** Use `task-master models` with `--set-<role>` and **also** `--ollama` or `--openrouter`.
+            -   Example: `task-master models --set-main=my-custom-ollama-model --ollama`
+            -   Example: `task-master models --set-main=some-openrouter-model --openrouter`
+        -   **Interactive Setup:** Use `task-master models --setup` and select the `Ollama (Enter Custom ID)` or `OpenRouter (Enter Custom ID)` options.
+    -   **OpenRouter Validation:** When setting a custom OpenRouter model, Taskmaster attempts to validate the ID against the live OpenRouter API.
+    -   **Ollama:** No live validation occurs for custom Ollama models; ensure the model is available on your Ollama server.
+
+-   **Supported Providers & Required API Keys:**
+    -   Task Master integrates with various providers via the Vercel AI SDK.
+    -   **API keys are essential** for most providers and must be configured correctly.
+    -   **Key Locations** (See [`dev_workflow.mdc`](mdc:.cursor/rules/dev_workflow.mdc) - Configuration Management):
+        -   **MCP/Cursor:** Set keys in the `env` section of `.cursor/mcp.json`.
+        -   **CLI:** Set keys in a `.env` file in the project root.
+    -   **Provider List & Keys:**
+        -   **`anthropic`**: Requires `ANTHROPIC_API_KEY`.
+        -   **`google`**: Requires `GOOGLE_API_KEY`.
+        -   **`openai`**: Requires `OPENAI_API_KEY`.
+        -   **`perplexity`**: Requires `PERPLEXITY_API_KEY`.
+        -   **`xai`**: Requires `XAI_API_KEY`.
+        -   **`mistral`**: Requires `MISTRAL_API_KEY`.
+        -   **`azure`**: Requires `AZURE_OPENAI_API_KEY` and `AZURE_OPENAI_ENDPOINT`.
+        -   **`openrouter`**: Requires `OPENROUTER_API_KEY`.
+        -   **`ollama`**: Might require `OLLAMA_API_KEY` (not currently supported) *and* `OLLAMA_BASE_URL` (default: `http://localhost:11434/api`). *Check specific setup.*
+
+-   **Troubleshooting:**
+    -   If AI commands fail (especially in MCP context):
+        1.  **Verify API Key:** Ensure the correct API key for the *selected provider* (check `models` output) exists in the appropriate location (`.cursor/mcp.json` env or `.env`).
+        2.  **Check Model ID:** Ensure the model ID set for the role is valid (use `models` listAvailableModels/`task-master models`).
+        3.  **Provider Status:** Check the status of the external AI provider's service.
+        4.  **Restart MCP:** If changes were made to configuration or provider code, restart the MCP server.
+
+## Adding a New AI Provider (Vercel AI SDK Method)
+
+Follow these steps to integrate a new AI provider that has an official Vercel AI SDK adapter (`@ai-sdk/<provider>`):
+
+1.  **Install Dependency:**
+    -   Install the provider-specific package:
+        ```bash
+        npm install @ai-sdk/<provider-name>
+        ```
+
+2.  **Create Provider Module:**
+    -   Create a new file in `src/ai-providers/` named `<provider-name>.js`.
+    -   Use existing modules (`openai.js`, `anthropic.js`, etc.) as a template.
+    -   **Import:**
+        -   Import the provider's `create<ProviderName>` function from `@ai-sdk/<provider-name>`.
+        -   Import `generateText`, `streamText`, `generateObject` from the core `ai` package.
+        -   Import the `log` utility from `../../scripts/modules/utils.js`.
+    -   **Implement Core Functions:**
+        -   `generate<ProviderName>Text(params)`:
+            -   Accepts `params` (apiKey, modelId, messages, etc.).
+            -   Instantiate the client: `const client = create<ProviderName>({ apiKey });`
+            -   Call `generateText({ model: client(modelId), ... })`.
+            -   Return `result.text`.
+            -   Include basic validation and try/catch error handling.
+        -   `stream<ProviderName>Text(params)`:
+            -   Similar structure to `generateText`.
+            -   Call `streamText({ model: client(modelId), ... })`.
+            -   Return the full stream result object.
+            -   Include basic validation and try/catch.
+        -   `generate<ProviderName>Object(params)`:
+            -   Similar structure.
+            -   Call `generateObject({ model: client(modelId), schema, messages, ... })`.
+            -   Return `result.object`.
+            -   Include basic validation and try/catch.
+    -   **Export Functions:** Export the three implemented functions (`generate<ProviderName>Text`, `stream<ProviderName>Text`, `generate<ProviderName>Object`).
+
+3.  **Integrate with Unified Service:**
+    -   Open `scripts/modules/ai-services-unified.js`.
+    -   **Import:** Add `import * as <providerName> from '../../src/ai-providers/<provider-name>.js';`
+    -   **Map:** Add an entry to the `PROVIDER_FUNCTIONS` map:
+        ```javascript
+        '<provider-name>': {
+            generateText: <providerName>.generate<ProviderName>Text,
+            streamText: <providerName>.stream<ProviderName>Text,
+            generateObject: <providerName>.generate<ProviderName>Object
+        },
+        ```
+
+4.  **Update Configuration Management:**
+    -   Open `scripts/modules/config-manager.js`.
+    -   **`MODEL_MAP`:** Add the new `<provider-name>` key to the `MODEL_MAP` loaded from `supported-models.json` (or ensure the loading handles new providers dynamically if `supported-models.json` is updated first).
+    -   **`VALID_PROVIDERS`:** Ensure the new `<provider-name>` is included in the `VALID_PROVIDERS` array (this should happen automatically if derived from `MODEL_MAP` keys).
+    -   **API Key Handling:**
+        -   Update the `keyMap` in `_resolveApiKey` and `isApiKeySet` with the correct environment variable name (e.g., `PROVIDER_API_KEY`).
+        -   Update the `switch` statement in `getMcpApiKeyStatus` to check the corresponding key in `mcp.json` and its placeholder value.
+        -   Add a case to the `switch` statement in `getMcpApiKeyStatus` for the new provider, including its placeholder string if applicable.
+    -   **Ollama Exception:** If adding Ollama or another provider *not* requiring an API key, add a specific check at the beginning of `isApiKeySet` and `getMcpApiKeyStatus` to return `true` immediately for that provider.
+
+5.  **Update Supported Models List:**
+    -   Edit `scripts/modules/supported-models.json`.
+    -   Add a new key for the `<provider-name>`.
+    -   Add an array of model objects under the provider key, each including:
+        -   `id`: The specific model identifier (e.g., `claude-3-opus-20240229`).
+        -   `name`: A user-friendly name (optional).
+        -   `swe_score`, `cost_per_1m_tokens`: (Optional) Add performance/cost data if available.
+        -   `allowed_roles`: An array of roles (`"main"`, `"research"`, `"fallback"`) the model is suitable for.
+        -   `max_tokens`: (Optional but recommended) The maximum token limit for the model.
+
+6.  **Update Environment Examples:**
+    -   Add the new `PROVIDER_API_KEY` to `.env.example`.
+    -   Add the new `PROVIDER_API_KEY` with its placeholder (`YOUR_PROVIDER_API_KEY_HERE`) to the `env` section for `taskmaster-ai` in `.cursor/mcp.json.example` (if it exists) or update instructions.
+
+7.  **Add Unit Tests:**
+    -   Create `tests/unit/ai-providers/<provider-name>.test.js`.
+    -   Mock the `@ai-sdk/<provider-name>` module and the core `ai` module functions (`generateText`, `streamText`, `generateObject`).
+    -   Write tests for each exported function (`generate<ProviderName>Text`, etc.) to verify:
+        -   Correct client instantiation.
+        -   Correct parameters passed to the mocked Vercel AI SDK functions.
+        -   Correct handling of results.
+        -   Error handling (missing API key, SDK errors).
+
+8.  **Documentation:**
+    -   Update any relevant documentation (like `README.md` or other rules) mentioning supported providers or configuration.
+
+*(Note: For providers **without** an official Vercel AI SDK adapter, the process would involve directly using the provider's own SDK or API within the `src/ai-providers/<provider-name>.js` module and manually constructing responses compatible with the unified service layer, which is significantly more complex.)*

.cursor/rules/ai_services.mdc

@@ -0,0 +1,101 @@
+---
+description: Guidelines for interacting with the unified AI service layer.
+globs: scripts/modules/ai-services-unified.js, scripts/modules/task-manager/*.js, scripts/modules/commands.js
+---
+
+# AI Services Layer Guidelines
+
+This document outlines the architecture and usage patterns for interacting with Large Language Models (LLMs) via Task Master's unified AI service layer (`ai-services-unified.js`). The goal is to centralize configuration, provider selection, API key management, fallback logic, and error handling.
+
+**Core Components:**
+
+*   **Configuration (`.taskmasterconfig` & [`config-manager.js`](mdc:scripts/modules/config-manager.js)):**
+    *   Defines the AI provider and model ID for different **roles** (`main`, `research`, `fallback`).
+    *   Stores parameters like `maxTokens` and `temperature` per role.
+    *   Managed via the `task-master models --setup` CLI command.
+    *   [`config-manager.js`](mdc:scripts/modules/config-manager.js) provides **getters** (e.g., `getMainProvider()`, `getParametersForRole()`) to access these settings. Core logic should **only** use these getters for *non-AI related application logic* (e.g., `getDefaultSubtasks`). The unified service fetches necessary AI parameters internally based on the `role`.
+    *   **API keys** are **NOT** stored here; they are resolved via `resolveEnvVariable` (in [`utils.js`](mdc:scripts/modules/utils.js)) from `.env` (for CLI) or the MCP `session.env` object (for MCP calls). See [`utilities.mdc`](mdc:.cursor/rules/utilities.mdc) and [`dev_workflow.mdc`](mdc:.cursor/rules/dev_workflow.mdc).
+
+*   **Unified Service (`ai-services-unified.js`):**
+    *   Exports primary interaction functions: `generateTextService`, `generateObjectService`. (Note: `streamTextService` exists but has known reliability issues with some providers/payloads).
+    *   Contains the core `_unifiedServiceRunner` logic.
+    *   Internally uses `config-manager.js` getters to determine the provider/model/parameters based on the requested `role`.
+    *   Implements the **fallback sequence** (e.g., main -> fallback -> research) if the primary provider/model fails.
+    *   Constructs the `messages` array required by the Vercel AI SDK.
+    *   Implements **retry logic** for specific API errors (`_attemptProviderCallWithRetries`).
+    *   Resolves API keys automatically via `_resolveApiKey` (using `resolveEnvVariable`).
+    *   Maps requests to the correct provider implementation (in `src/ai-providers/`) via `PROVIDER_FUNCTIONS`.
+
+*   **Provider Implementations (`src/ai-providers/*.js`):**
+    *   Contain provider-specific wrappers around Vercel AI SDK functions (`generateText`, `generateObject`).
+
+**Usage Pattern (from Core Logic like `task-manager/*.js`):**
+
+1.  **Import Service:** Import `generateTextService` or `generateObjectService` from `../ai-services-unified.js`.
+    ```javascript
+    // Preferred for most tasks (especially with complex JSON)
+    import { generateTextService } from '../ai-services-unified.js';
+
+    // Use if structured output is reliable for the specific use case
+    // import { generateObjectService } from '../ai-services-unified.js';
+    ```
+
+2.  **Prepare Parameters:** Construct the parameters object for the service call.
+    *   `role`: **Required.** `'main'`, `'research'`, or `'fallback'`. Determines the initial provider/model/parameters used by the unified service.
+    *   `session`: **Required if called from MCP context.** Pass the `session` object received by the direct function wrapper. The unified service uses `session.env` to find API keys.
+    *   `systemPrompt`: Your system instruction string.
+    *   `prompt`: The user message string (can be long, include stringified data, etc.).
+    *   (For `generateObjectService` only): `schema` (Zod schema), `objectName`.
+
+3.  **Call Service:** Use `await` to call the service function.
+    ```javascript
+    // Example using generateTextService (most common)
+    try {
+        const resultText = await generateTextService({
+            role: useResearch ? 'research' : 'main', // Determine role based on logic
+            session: context.session, // Pass session from context object
+            systemPrompt: "You are...",
+            prompt: userMessageContent
+        });
+        // Process the raw text response (e.g., parse JSON, use directly)
+        // ...
+    } catch (error) {
+        // Handle errors thrown by the unified service (if all fallbacks/retries fail)
+        report('error', `Unified AI service call failed: ${error.message}`);
+        throw error;
+    }
+
+    // Example using generateObjectService (use cautiously)
+    try {
+        const resultObject = await generateObjectService({
+            role: 'main',
+            session: context.session,
+            schema: myZodSchema,
+            objectName: 'myDataObject',
+            systemPrompt: "You are...",
+            prompt: userMessageContent
+        });
+        // resultObject is already a validated JS object
+        // ...
+    } catch (error) {
+        report('error', `Unified AI service call failed: ${error.message}`);
+        throw error;
+    }
+    ```
+
+4.  **Handle Results/Errors:** Process the returned text/object or handle errors thrown by the unified service layer.
+
+**Key Implementation Rules & Gotchas:**
+
+*   ✅ **DO**: Centralize **all** LLM calls through `generateTextService` or `generateObjectService`.
+*   ✅ **DO**: Determine the appropriate `role` (`main`, `research`, `fallback`) in your core logic and pass it to the service.
+*   ✅ **DO**: Pass the `session` object (received in the `context` parameter, especially from direct function wrappers) to the service call when in MCP context.
+*   ✅ **DO**: Ensure API keys are correctly configured in `.env` (for CLI) or `.cursor/mcp.json` (for MCP).
+*   ✅ **DO**: Ensure `.taskmasterconfig` exists and has valid provider/model IDs for the roles you intend to use (manage via `task-master models --setup`).
+*   ✅ **DO**: Use `generateTextService` and implement robust manual JSON parsing (with Zod validation *after* parsing) when structured output is needed, as `generateObjectService` has shown unreliability with some providers/schemas.
+*   ❌ **DON'T**: Import or call anything from the old `ai-services.js`, `ai-client-factory.js`, or `ai-client-utils.js` files.
+*   ❌ **DON'T**: Initialize AI clients (Anthropic, Perplexity, etc.) directly within core logic (`task-manager/`) or MCP direct functions.
+*   ❌ **DON'T**: Fetch AI-specific parameters (model ID, max tokens, temp) using `config-manager.js` getters *for the AI call*. Pass the `role` instead.
+*   ❌ **DON'T**: Implement fallback or retry logic outside `ai-services-unified.js`.
+*   ❌ **DON'T**: Handle API key resolution outside the service layer (it uses `utils.js` internally).
+*   ⚠️ **generateObjectService Caution**: Be aware of potential reliability issues with `generateObjectService` across different providers and complex schemas. Prefer `generateTextService` + manual parsing as a more robust alternative for structured data needs.

.cursor/rules/architecture.mdc

@@ -3,7 +3,6 @@ description: Describes the high-level architecture of the Task Master CLI applic
 globs: scripts/modules/*.js
 alwaysApply: false
 ---
-
 # Application Architecture Overview
 
 - **Modular Structure**: The Task Master CLI is built using a modular architecture, with distinct modules responsible for different aspects of the application. This promotes separation of concerns, maintainability, and testability.
@@ -14,161 +13,74 @@ alwaysApply: false
     - **Purpose**: Defines and registers all CLI commands using Commander.js.
     - **Responsibilities** (See also: [`commands.mdc`](mdc:.cursor/rules/commands.mdc)):
       - Parses command-line arguments and options.
-      - Invokes appropriate functions from other modules to execute commands (e.g., calls `initializeProject` from `init.js` for the `init` command).
-      - Handles user input and output related to command execution.
-      - Implements input validation and error handling for CLI commands.
-    - **Key Components**:
-      - `programInstance` (Commander.js `Command` instance): Manages command definitions.
-      - `registerCommands(programInstance)`: Function to register all application commands.
-      - Command action handlers: Functions executed when a specific command is invoked, delegating to core modules.
+      - Invokes appropriate core logic functions from `scripts/modules/`.
+      - Handles user input/output for CLI.
+      - Implements CLI-specific validation.
 
-  - **[`task-manager.js`](mdc:scripts/modules/task-manager.js): Task Data Management**
-    - **Purpose**: Manages task data, including loading, saving, creating, updating, deleting, and querying tasks.
+  - **[`task-manager.js`](mdc:scripts/modules/task-manager.js) & `task-manager/` directory: Task Data & Core Logic**
+    - **Purpose**: Contains core functions for task data manipulation (CRUD), AI interactions, and related logic.
     - **Responsibilities**:
-      - Reads and writes task data to `tasks.json` file.
-      - Implements functions for task CRUD operations (Create, Read, Update, Delete).
-      - Handles task parsing from PRD documents using AI.
-      - Manages task expansion and subtask generation.
-      - Updates task statuses and properties.
-      - Implements task listing and display logic.
-      - Performs task complexity analysis using AI.
-    - **Key Functions**:
-      - `readTasks(tasksPath)` / `writeTasks(tasksPath, tasksData)`: Load and save task data.
-      - `parsePRD(prdFilePath, outputPath, numTasks)`: Parses PRD document to create tasks.
-      - `expandTask(taskId, numSubtasks, useResearch, prompt, force)`: Expands a task into subtasks.
-      - `setTaskStatus(tasksPath, taskIdInput, newStatus)`: Updates task status.
-      - `listTasks(tasksPath, statusFilter, withSubtasks)`: Lists tasks with filtering and subtask display options.
-      - `analyzeComplexity(tasksPath, reportPath, useResearch, thresholdScore)`: Analyzes task complexity.
+      - Reading/writing `tasks.json`.
+      - Implementing functions for task CRUD, parsing PRDs, expanding tasks, updating status, etc.
+      - **Delegating AI interactions** to the `ai-services-unified.js` layer.
+      - Accessing non-AI configuration via `config-manager.js` getters.
+    - **Key Files**: Individual files within `scripts/modules/task-manager/` handle specific actions (e.g., `add-task.js`, `expand-task.js`).
 
   - **[`dependency-manager.js`](mdc:scripts/modules/dependency-manager.js): Dependency Management**
-    - **Purpose**: Manages task dependencies, including adding, removing, validating, and fixing dependency relationships.
-    - **Responsibilities**:
-      - Adds and removes task dependencies.
-      - Validates dependency relationships to prevent circular dependencies and invalid references.
-      - Fixes invalid dependencies by removing non-existent or self-referential dependencies.
-      - Provides functions to check for circular dependencies.
-    - **Key Functions**:
-      - `addDependency(tasksPath, taskId, dependencyId)`: Adds a dependency between tasks.
-      - `removeDependency(tasksPath, taskId, dependencyId)`: Removes a dependency.
-      - `validateDependencies(tasksPath)`: Validates task dependencies.
-      - `fixDependencies(tasksPath)`: Fixes invalid task dependencies.
-      - `isCircularDependency(tasks, taskId, dependencyChain)`: Detects circular dependencies.
+    - **Purpose**: Manages task dependencies.
+    - **Responsibilities**: Add/remove/validate/fix dependencies.
 
   - **[`ui.js`](mdc:scripts/modules/ui.js): User Interface Components**
-    - **Purpose**: Handles all user interface elements, including displaying information, formatting output, and providing user feedback.
-    - **Responsibilities**:
-      - Displays task lists, task details, and command outputs in a formatted way.
-      - Uses `chalk` for colored output and `boxen` for boxed messages.
-      - Implements table display using `cli-table3`.
-      - Shows loading indicators using `ora`.
-      - Provides helper functions for status formatting, dependency display, and progress reporting.
-      - Suggests next actions to the user after command execution.
-    - **Key Functions**:
-      - `displayTaskList(tasks, statusFilter, withSubtasks)`: Displays a list of tasks in a table.
-      - `displayTaskDetails(task)`: Displays detailed information for a single task.
-      - `displayComplexityReport(reportPath)`: Displays the task complexity report.
-      - `startLoadingIndicator(message)` / `stopLoadingIndicator(indicator)`: Manages loading indicators.
-      - `getStatusWithColor(status)`: Returns status string with color formatting.
-      - `formatDependenciesWithStatus(dependencies, allTasks, inTable)`: Formats dependency list with status indicators.
+    - **Purpose**: Handles CLI output formatting (tables, colors, boxes, spinners).
+    - **Responsibilities**: Displaying tasks, reports, progress, suggestions.
 
-  - **[`ai-services.js`](mdc:scripts/modules/ai-services.js) (Conceptual): AI Integration**
-    - **Purpose**:  Abstracts interactions with AI models (like Anthropic Claude and Perplexity AI) for various features. *Note: This module might be implicitly implemented within `task-manager.js` and `utils.js` or could be explicitly created for better organization as the project evolves.*
-    - **Responsibilities**:
-      - Handles API calls to AI services.
-      - Manages prompts and parameters for AI requests.
-      - Parses AI responses and extracts relevant information.
-      - Implements logic for task complexity analysis, task expansion, and PRD parsing using AI.
-    - **Potential Functions**:
-      - `getAIResponse(prompt, model, maxTokens, temperature)`: Generic function to interact with AI model.
-      - `analyzeTaskComplexityWithAI(taskDescription)`: Sends task description to AI for complexity analysis.
-      - `expandTaskWithAI(taskDescription, numSubtasks, researchContext)`: Generates subtasks using AI.
-      - `parsePRDWithAI(prdContent)`: Extracts tasks from PRD content using AI.
+  - **[`ai-services-unified.js`](mdc:scripts/modules/ai-services-unified.js): Unified AI Service Layer**
+    - **Purpose**: Centralized interface for all LLM interactions using Vercel AI SDK.
+    - **Responsibilities** (See also: [`ai_services.mdc`](mdc:.cursor/rules/ai_services.mdc)):
+      - Exports `generateTextService`, `generateObjectService`.
+      - Handles provider/model selection based on `role` and `.taskmasterconfig`.
+      - Resolves API keys (from `.env` or `session.env`).
+      - Implements fallback and retry logic.
+      - Orchestrates calls to provider-specific implementations (`src/ai-providers/`).
 
-  - **[`utils.js`](mdc:scripts/modules/utils.js): Utility Functions and Configuration**
-    - **Purpose**: Provides reusable utility functions and global configuration settings used across the **CLI application**.
+  - **[`src/ai-providers/*.js`](mdc:src/ai-providers/): Provider-Specific Implementations**
+    - **Purpose**: Provider-specific wrappers for Vercel AI SDK functions.
+    - **Responsibilities**: Interact directly with Vercel AI SDK adapters.
+
+  - **[`config-manager.js`](mdc:scripts/modules/config-manager.js): Configuration Management**
+    - **Purpose**: Loads, validates, and provides access to configuration.
     - **Responsibilities** (See also: [`utilities.mdc`](mdc:.cursor/rules/utilities.mdc)):
-      - Manages global configuration settings loaded from environment variables and defaults.
-      - Implements logging utility with different log levels and output formatting.
-      - Provides file system operation utilities (read/write JSON files).
-      - Includes string manipulation utilities (e.g., `truncate`, `sanitizePrompt`).
-      - Offers task-specific utility functions (e.g., `formatTaskId`, `findTaskById`, `taskExists`).
-      - Implements graph algorithms like cycle detection for dependency management.
-      - **Silent Mode Control**: Provides `enableSilentMode` and `disableSilentMode` functions to control log output.
-    - **Key Components**:
-      - `CONFIG`: Global configuration object.
-      - `log(level, ...args)`: Logging function.
-      - `readJSON(filepath)` / `writeJSON(filepath, data)`: File I/O utilities for JSON files.
-      - `truncate(text, maxLength)`: String truncation utility.
-      - `formatTaskId(id)` / `findTaskById(tasks, taskId)`: Task ID and search utilities.
-      - `findCycles(subtaskId, dependencyMap)`: Cycle detection algorithm.
-      - `enableSilentMode()` / `disableSilentMode()`: Control console logging output.
+      - Reads and merges `.taskmasterconfig` with defaults.
+      - Provides getters (e.g., `getMainProvider`, `getLogLevel`, `getDefaultSubtasks`) for accessing settings.
+      - **Note**: Does **not** store or directly handle API keys (keys are in `.env` or MCP `session.env`).
+
+  - **[`utils.js`](mdc:scripts/modules/utils.js): Core Utility Functions**
+    - **Purpose**: Low-level, reusable CLI utilities.
+    - **Responsibilities** (See also: [`utilities.mdc`](mdc:.cursor/rules/utilities.mdc)):
+      - Logging (`log` function), File I/O (`readJSON`, `writeJSON`), String utils (`truncate`).
+      - Task utils (`findTaskById`), Dependency utils (`findCycles`).
+      - API Key Resolution (`resolveEnvVariable`).
+      - Silent Mode Control (`enableSilentMode`, `disableSilentMode`).
 
   - **[`mcp-server/`](mdc:mcp-server/): MCP Server Integration**
-    - **Purpose**: Provides an MCP (Model Context Protocol) interface for Task Master, allowing integration with external tools like Cursor. Uses FastMCP framework.
+    - **Purpose**: Provides MCP interface using FastMCP.
     - **Responsibilities** (See also: [`mcp.mdc`](mdc:.cursor/rules/mcp.mdc)):
-      - Registers Task Master functionalities as tools consumable via MCP.
-      - Handles MCP requests via tool `execute` methods defined in `mcp-server/src/tools/*.js`.
-      - Tool `execute` methods call corresponding **direct function wrappers**.
-      - Tool `execute` methods use `getProjectRootFromSession` (from [`tools/utils.js`](mdc:mcp-server/src/tools/utils.js)) to determine the project root from the client session and pass it to the direct function.
-      - **Direct function wrappers (`*Direct` functions in `mcp-server/src/core/direct-functions/*.js`) contain the main logic for handling MCP requests**, including path resolution, argument validation, caching, and calling core Task Master functions.
-      - Direct functions use `findTasksJsonPath` (from [`core/utils/path-utils.js`](mdc:mcp-server/src/core/utils/path-utils.js)) to locate `tasks.json` based on the provided `projectRoot`.
-      - **Silent Mode Implementation**: Direct functions use `enableSilentMode` and `disableSilentMode` to prevent logs from interfering with JSON responses.
-      - **Async Operations**: Uses `AsyncOperationManager` to handle long-running operations in the background.
-      - **Project Initialization**: Provides `initialize_project` command for setting up new projects from within integrated clients.
-      - Tool `execute` methods use `handleApiResult` from [`tools/utils.js`](mdc:mcp-server/src/tools/utils.js) to process the result from the direct function and format the final MCP response.
-      - Uses CLI execution via `executeTaskMasterCommand` as a fallback only when necessary.
-      - **Implements Robust Path Finding**: The utility [`tools/utils.js`](mdc:mcp-server/src/tools/utils.js) (specifically `getProjectRootFromSession`) and [`core/utils/path-utils.js`](mdc:mcp-server/src/core/utils/path-utils.js) (specifically `findTasksJsonPath`) work together. The tool gets the root via session, passes it to the direct function, which uses `findTasksJsonPath` to locate the specific `tasks.json` file within that root.
-      - **Implements Caching**: Utilizes a caching layer (`ContextManager` with `lru-cache`). Caching logic is invoked *within* the direct function wrappers using the `getCachedOrExecute` utility for performance-sensitive read operations.
-      - Standardizes response formatting and data filtering using utilities in [`tools/utils.js`](mdc:mcp-server/src/tools/utils.js).
-      - **Resource Management**: Provides access to static and dynamic resources.
-    - **Key Components**:
-      - `mcp-server/src/index.js`: Main server class definition with FastMCP initialization, resource registration, and server lifecycle management.
-      - `mcp-server/src/server.js`: Main server setup and initialization.
-      - `mcp-server/src/tools/`: Directory containing individual tool definitions. Each tool's `execute` method orchestrates the call to core logic and handles the response.
-      - `mcp-server/src/tools/utils.js`: Provides MCP-specific utilities like `handleApiResult`, `processMCPResponseData`, `getCachedOrExecute`, and **`getProjectRootFromSession`**.
-      - `mcp-server/src/core/utils/`: Directory containing utility functions specific to the MCP server, like **`path-utils.js` for resolving `tasks.json` within a given root** and **`async-manager.js` for handling background operations**.
-      - `mcp-server/src/core/direct-functions/`: Directory containing individual files for each **direct function wrapper (`*Direct`)**. These files contain the primary logic for MCP tool execution.
-      - `mcp-server/src/core/resources/`: Directory containing resource handlers for task templates, workflow definitions, and other static/dynamic data exposed to LLM clients.
-      - [`task-master-core.js`](mdc:mcp-server/src/core/task-master-core.js): Acts as an import/export hub, collecting and exporting direct functions from the `direct-functions` directory and MCP utility functions.
-    - **Naming Conventions**:
-      - **Files** use **kebab-case**: `list-tasks.js`, `set-task-status.js`, `parse-prd.js`
-      - **Direct Functions** use **camelCase** with `Direct` suffix: `listTasksDirect`, `setTaskStatusDirect`, `parsePRDDirect`
-      - **Tool Registration Functions** use **camelCase** with `Tool` suffix: `registerListTasksTool`, `registerSetTaskStatusTool`
-      - **MCP Tool Names** use **snake_case**: `list_tasks`, `set_task_status`, `parse_prd_document`
-      - **Resource Handlers** use **camelCase** with pattern URI: `@mcp.resource("tasks://templates/{template_id}")`
-    - **AsyncOperationManager**:
-      - **Purpose**: Manages background execution of long-running operations.
-      - **Location**: `mcp-server/src/core/utils/async-manager.js`
-      - **Key Features**:
-        - Operation tracking with unique IDs using UUID
-        - Status management (pending, running, completed, failed)
-        - Progress reporting forwarded from background tasks
-        - Operation history with automatic cleanup of completed operations
-        - Context preservation (log, session, reportProgress)
-        - Robust error handling for background tasks
-      - **Usage**: Used for CPU-intensive operations like task expansion and PRD parsing
+      - Registers tools (`mcp-server/src/tools/*.js`). Tool `execute` methods **should be wrapped** with the `withNormalizedProjectRoot` HOF (from `tools/utils.js`) to ensure consistent path handling.
+      - The HOF provides a normalized `args.projectRoot` to the `execute` method.
+      - Tool `execute` methods call **direct function wrappers** (`mcp-server/src/core/direct-functions/*.js`), passing the normalized `projectRoot` and other args.
+      - Direct functions use path utilities (`mcp-server/src/core/utils/`) to resolve paths based on `projectRoot` from session.
+      - Direct functions implement silent mode, logger wrappers, and call core logic functions from `scripts/modules/`.
+      - Manages MCP caching and response formatting.
 
   - **[`init.js`](mdc:scripts/init.js): Project Initialization Logic**
-    - **Purpose**: Contains the core logic for setting up a new Task Master project structure.
-    - **Responsibilities**:
-      - Creates necessary directories (`.cursor/rules`, `scripts`, `tasks`).
-      - Copies template files (`.env.example`, `.gitignore`, rule files, `dev.js`, etc.).
-      - Creates or merges `package.json` with required dependencies and scripts.
-      - Sets up MCP configuration (`.cursor/mcp.json`).
-      - Optionally initializes a git repository and installs dependencies.
-      - Handles user prompts for project details *if* called without skip flags (`-y`).
-    - **Key Function**:
-      - `initializeProject(options)`: The main function exported and called by the `init` command's action handler in [`commands.js`](mdc:scripts/modules/commands.js). It receives parsed options directly.
-    - **Note**: This script is used as a module and no longer handles its own argument parsing or direct execution via a separate `bin` file.
+    - **Purpose**: Sets up new Task Master project structure.
+    - **Responsibilities**: Creates directories, copies templates, manages `package.json`, sets up `.cursor/mcp.json`.
 
-- **Data Flow and Module Dependencies**:
+- **Data Flow and Module Dependencies (Updated)**:
 
-  - **Commands Initiate Actions**: User commands entered via the CLI (parsed by `commander` based on definitions in [`commands.js`](mdc:scripts/modules/commands.js)) are the entry points for most operations.
-  - **Command Handlers Delegate to Core Logic**: Action handlers within [`commands.js`](mdc:scripts/modules/commands.js) call functions in core modules like [`task-manager.js`](mdc:scripts/modules/task-manager.js), [`dependency-manager.js`](mdc:scripts/modules/dependency-manager.js), and [`init.js`](mdc:scripts/init.js) (for the `init` command) to perform the actual work.
-  - **UI for Presentation**:  [`ui.js`](mdc:scripts/modules/ui.js) is used by command handlers and task/dependency managers to display information to the user. UI functions primarily consume data and format it for output, without modifying core application state.
-  - **Utilities for Common Tasks**: [`utils.js`](mdc:scripts/modules/utils.js) provides helper functions used by all other modules for configuration, logging, file operations, and common data manipulations.
-  - **AI Services Integration**: AI functionalities (complexity analysis, task expansion, PRD parsing) are invoked from [`task-manager.js`](mdc:scripts/modules/task-manager.js) and potentially [`commands.js`](mdc:scripts/modules/commands.js), likely using functions that would reside in a dedicated `ai-services.js` module or be integrated within `utils.js` or `task-manager.js`.
-  - **MCP Server Interaction**: External tools interact with the `mcp-server`. MCP Tool `execute` methods use `getProjectRootFromSession` to find the project root, then call direct function wrappers (in `mcp-server/src/core/direct-functions/`) passing the root in `args`. These wrappers handle path finding for `tasks.json` (using `path-utils.js`), validation, caching, call the core logic from `scripts/modules/` (passing logging context via the standard wrapper pattern detailed in mcp.mdc), and return a standardized result. The final MCP response is formatted by `mcp-server/src/tools/utils.js`. See [`mcp.mdc`](mdc:.cursor/rules/mcp.mdc) for details.
+  - **CLI**: `bin/task-master.js` -> `scripts/dev.js` (loads `.env`) -> `scripts/modules/commands.js` -> Core Logic (`scripts/modules/*`) -> Unified AI Service (`ai-services-unified.js`) -> Provider Adapters -> LLM API.
+  - **MCP**: External Tool -> `mcp-server/server.js` -> Tool (`mcp-server/src/tools/*`) -> Direct Function (`mcp-server/src/core/direct-functions/*`) -> Core Logic (`scripts/modules/*`) -> Unified AI Service (`ai-services-unified.js`) -> Provider Adapters -> LLM API.
+  - **Configuration**: Core logic needing non-AI settings calls `config-manager.js` getters (passing `session.env` via `explicitRoot` if from MCP). Unified AI Service internally calls `config-manager.js` getters (using `role`) for AI params and `utils.js` (`resolveEnvVariable` with `session.env`) for API keys.
 
 ## Silent Mode Implementation Pattern in MCP Direct Functions
 
@@ -366,19 +278,8 @@ The `initialize_project` command provides a way to set up a new Task Master proj
   - Configures project metadata (name, description, version)
   - Handles shell alias creation if requested
   - Works in both interactive and non-interactive modes
-
-## Async Operation Management
-
-The AsyncOperationManager provides background task execution capabilities:
-
-- **Location**: `mcp-server/src/core/utils/async-manager.js`
-- **Key Components**:
-  - `asyncOperationManager` singleton instance
-  - `addOperation(operationFn, args, context)` method
-  - `getStatus(operationId)` method
-- **Usage Flow**:
-  1. Client calls an MCP tool that may take time to complete
-  2. Tool uses AsyncOperationManager to run the operation in background
-  3. Tool returns immediate response with operation ID
-  4. Client polls `get_operation_status` tool with the ID
-  5. Once completed, client can access operation results
+  - Creates necessary directories and files for a new project
+  - Sets up `tasks.json` and initial task files
+  - Configures project metadata (name, description, version)
+  - Handles shell alias creation if requested
+  - Works in both interactive and non-interactive modes

.cursor/rules/commands.mdc

@@ -34,8 +34,8 @@ While this document details the implementation of Task Master's **CLI commands**
 - **Command Handler Organization**:
   - ✅ DO: Keep action handlers concise and focused
   - ✅ DO: Extract core functionality to appropriate modules
-  - ✅ DO: Have the action handler import and call the relevant function(s) from core modules (e.g., `task-manager.js`, `init.js`), passing the parsed `options`.
-  - ✅ DO: Perform basic parameter validation (e.g., checking for required options) within the action handler or at the start of the called core function.
+  - ✅ DO: Have the action handler import and call the relevant functions from core modules, like `task-manager.js` or `init.js`, passing the parsed `options`.
+  - ✅ DO: Perform basic parameter validation, such as checking for required options, within the action handler or at the start of the called core function.
   - ❌ DON'T: Implement business logic in command handlers
 
 ## Best Practices for Removal/Delete Commands
@@ -44,7 +44,7 @@ When implementing commands that delete or remove data (like `remove-task` or `re
 
 - **Confirmation Prompts**:
   - ✅ **DO**: Include a confirmation prompt by default for destructive operations
-  - ✅ **DO**: Provide a `--yes` or `-y` flag to skip confirmation for scripting/automation
+  - ✅ **DO**: Provide a `--yes` or `-y` flag to skip confirmation, useful for scripting or automation
   - ✅ **DO**: Show what will be deleted in the confirmation message
   - ❌ **DON'T**: Perform destructive operations without user confirmation unless explicitly overridden
 
@@ -78,7 +78,7 @@ When implementing commands that delete or remove data (like `remove-task` or `re
 
 - **File Path Handling**:
   - ✅ **DO**: Use `path.join()` to construct file paths
-  - ✅ **DO**: Follow established naming conventions for tasks (e.g., `task_001.txt`)
+  - ✅ **DO**: Follow established naming conventions for tasks, like `task_001.txt`
   - ✅ **DO**: Check if files exist before attempting to delete them
   - ✅ **DO**: Handle file deletion errors gracefully
   - ❌ **DON'T**: Construct paths with string concatenation
@@ -166,10 +166,10 @@ When implementing commands that delete or remove data (like `remove-task` or `re
   - ✅ DO: Use descriptive, action-oriented names
 
 - **Option Names**:
-  - ✅ DO: Use kebab-case for long-form option names (`--output-format`)
-  - ✅ DO: Provide single-letter shortcuts when appropriate (`-f, --file`)
+  - ✅ DO: Use kebab-case for long-form option names, like `--output-format`
+  - ✅ DO: Provide single-letter shortcuts when appropriate, like `-f, --file`
   - ✅ DO: Use consistent option names across similar commands
-  - ❌ DON'T: Use different names for the same concept (`--file` in one command, `--path` in another)
+  - ❌ DON'T: Use different names for the same concept, such as `--file` in one command and `--path` in another
 
   ```javascript
   // ✅ DO: Use consistent option naming
@@ -181,7 +181,7 @@ When implementing commands that delete or remove data (like `remove-task` or `re
   .option('-p, --path <dir>', 'Output directory') // Should be --output
   ```
 
-  > **Note**: Although options are defined with kebab-case (`--num-tasks`), Commander.js stores them internally as camelCase properties. Access them in code as `options.numTasks`, not `options['num-tasks']`.
+  > **Note**: Although options are defined with kebab-case, like `--num-tasks`, Commander.js stores them internally as camelCase properties. Access them in code as `options.numTasks`, not `options['num-tasks']`.
 
 - **Boolean Flag Conventions**:
   - ✅ DO: Use positive flags with `--skip-` prefix for disabling behavior
@@ -210,7 +210,7 @@ When implementing commands that delete or remove data (like `remove-task` or `re
 - **Required Parameters**:
   - ✅ DO: Check that required parameters are provided
   - ✅ DO: Provide clear error messages when parameters are missing
-  - ✅ DO: Use early returns with process.exit(1) for validation failures
+  - ✅ DO: Use early returns with `process.exit(1)` for validation failures
 
   ```javascript
   // ✅ DO: Validate required parameters early
@@ -221,7 +221,7 @@ When implementing commands that delete or remove data (like `remove-task` or `re
   ```
 
 - **Parameter Type Conversion**:
-  - ✅ DO: Convert string inputs to appropriate types (numbers, booleans)
+  - ✅ DO: Convert string inputs to appropriate types, such as numbers or booleans
   - ✅ DO: Handle conversion errors gracefully
 
   ```javascript
@@ -254,7 +254,7 @@ When implementing commands that delete or remove data (like `remove-task` or `re
   const taskId = parseInt(options.id, 10);
   if (isNaN(taskId) || taskId <= 0) {
     console.error(chalk.red(`Error: Invalid task ID: ${options.id}. Task ID must be a positive integer.`));
-    console.log(chalk.yellow('Usage example: task-master update-task --id=\'23\' --prompt=\'Update with new information.\nEnsure proper error handling.\''));
+    console.log(chalk.yellow("Usage example: task-master update-task --id='23' --prompt='Update with new information.\\nEnsure proper error handling.'"));
     process.exit(1);
   }
   
@@ -392,9 +392,9 @@ When implementing commands that delete or remove data (like `remove-task` or `re
   process.on('uncaughtException', (err) => {
     // Handle Commander-specific errors
     if (err.code === 'commander.unknownOption') {
-      const option = err.message.match(/'([^']+)'/)?.[1];
+      const option = err.message.match(/'([^']+)'/)?.[1]; // Safely extract option name
       console.error(chalk.red(`Error: Unknown option '${option}'`));
-      console.error(chalk.yellow(`Run 'task-master <command> --help' to see available options`));
+      console.error(chalk.yellow("Run 'task-master <command> --help' to see available options"));
       process.exit(1);
     }
     
@@ -464,9 +464,9 @@ When implementing commands that delete or remove data (like `remove-task` or `re
     .option('-f, --file <path>', 'Path to the tasks file', 'tasks/tasks.json')
     .option('-p, --parent <id>', 'ID of the parent task (required)')
     .option('-i, --task-id <id>', 'Existing task ID to convert to subtask')
-    .option('-t, --title <title>', 'Title for the new subtask (when not converting)')
-    .option('-d, --description <description>', 'Description for the new subtask (when not converting)')
-    .option('--details <details>', 'Implementation details for the new subtask (when not converting)')
+    .option('-t, --title <title>', 'Title for the new subtask, required if not converting')
+    .option('-d, --description <description>', 'Description for the new subtask, optional')
+    .option('--details <details>', 'Implementation details for the new subtask, optional')
     .option('--dependencies <ids>', 'Comma-separated list of subtask IDs this subtask depends on')
     .option('--status <status>', 'Initial status for the subtask', 'pending')
     .option('--skip-generate', 'Skip regenerating task files')
@@ -489,8 +489,8 @@ When implementing commands that delete or remove data (like `remove-task` or `re
     .command('remove-subtask')
     .description('Remove a subtask from its parent task, optionally converting it to a standalone task')
     .option('-f, --file <path>', 'Path to the tasks file', 'tasks/tasks.json')
-    .option('-i, --id <id>', 'ID of the subtask to remove in format "parentId.subtaskId" (required)')
-    .option('-c, --convert', 'Convert the subtask to a standalone task')
+    .option('-i, --id <id>', 'ID of the subtask to remove in format parentId.subtaskId, required')
+    .option('-c, --convert', 'Convert the subtask to a standalone task instead of deleting')
     .option('--skip-generate', 'Skip regenerating task files')
     .action(async (options) => {
       // Implementation with detailed error handling
@@ -513,7 +513,8 @@ When implementing commands that delete or remove data (like `remove-task` or `re
   // ✅ DO: Implement version checking function
   async function checkForUpdate() {
     // Implementation details...
-    return { currentVersion, latestVersion, needsUpdate };
+    // Example return structure:
+    return { currentVersion, latestVersion, updateAvailable };
   }
   
   // ✅ DO: Implement semantic version comparison
@@ -553,7 +554,7 @@ When implementing commands that delete or remove data (like `remove-task` or `re
       
       // After command execution, check if an update is available
       const updateInfo = await updateCheckPromise;
-      if (updateInfo.needsUpdate) {
+      if (updateInfo.updateAvailable) {
         displayUpgradeNotification(updateInfo.currentVersion, updateInfo.latestVersion);
       }
     } catch (error) {

.cursor/rules/dev_workflow.mdc

@@ -3,7 +3,6 @@ description: Guide for using Task Master to manage task-driven development workf
 globs: **/*
 alwaysApply: true
 ---
-
 # Task Master Development Workflow
 
 This guide outlines the typical process for using Task Master to manage software development projects.
@@ -29,21 +28,21 @@ Task Master offers two primary ways to interact:
 
 ## Standard Development Workflow Process
 
--   Start new projects by running `init` tool / `task-master init` or `parse_prd` / `task-master parse-prd --input='<prd-file.txt>'` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) to generate initial tasks.json
+-   Start new projects by running `initialize_project` tool / `task-master init` or `parse_prd` / `task-master parse-prd --input='<prd-file.txt>'` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) to generate initial tasks.json
 -   Begin coding sessions with `get_tasks` / `task-master list` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) to see current tasks, status, and IDs
 -   Determine the next task to work on using `next_task` / `task-master next` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)).
--   Analyze task complexity with `analyze_complexity` / `task-master analyze-complexity --research` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) before breaking down tasks
+-   Analyze task complexity with `analyze_project_complexity` / `task-master analyze-complexity --research` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) before breaking down tasks
 -   Review complexity report using `complexity_report` / `task-master complexity-report` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)).
 -   Select tasks based on dependencies (all marked 'done'), priority level, and ID order
 -   Clarify tasks by checking task files in tasks/ directory or asking for user input
 -   View specific task details using `get_task` / `task-master show <id>` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) to understand implementation requirements
--   Break down complex tasks using `expand_task` / `task-master expand --id=<id>` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) with appropriate flags
+-   Break down complex tasks using `expand_task` / `task-master expand --id=<id> --force --research` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) with appropriate flags like `--force` (to replace existing subtasks) and `--research`.
 -   Clear existing subtasks if needed using `clear_subtasks` / `task-master clear-subtasks --id=<id>` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) before regenerating
 -   Implement code following task details, dependencies, and project standards
 -   Verify tasks according to test strategies before marking as complete (See [`tests.mdc`](mdc:.cursor/rules/tests.mdc))
 -   Mark completed tasks with `set_task_status` / `task-master set-status --id=<id> --status=done` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc))
 -   Update dependent tasks when implementation differs from original plan using `update` / `task-master update --from=<id> --prompt="..."` or `update_task` / `task-master update-task --id=<id> --prompt="..."` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc))
--   Add new tasks discovered during implementation using `add_task` / `task-master add-task --prompt="..."` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)).
+-   Add new tasks discovered during implementation using `add_task` / `task-master add-task --prompt="..." --research` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)).
 -   Add new subtasks as needed using `add_subtask` / `task-master add-subtask --parent=<id> --title="..."` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)).
 -   Append notes or details to subtasks using `update_subtask` / `task-master update-subtask --id=<subtaskId> --prompt='Add implementation notes here...\nMore details...'` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)).
 -   Generate task files with `generate` / `task-master generate` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) after updating tasks.json
@@ -53,29 +52,30 @@ Task Master offers two primary ways to interact:
 
 ## Task Complexity Analysis
 
--   Run `analyze_complexity` / `task-master analyze-complexity --research` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) for comprehensive analysis
+-   Run `analyze_project_complexity` / `task-master analyze-complexity --research` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) for comprehensive analysis
 -   Review complexity report via `complexity_report` / `task-master complexity-report` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) for a formatted, readable version.
 -   Focus on tasks with highest complexity scores (8-10) for detailed breakdown
 -   Use analysis results to determine appropriate subtask allocation
--   Note that reports are automatically used by the `expand` tool/command
+-   Note that reports are automatically used by the `expand_task` tool/command
 
 ## Task Breakdown Process
 
--   For tasks with complexity analysis, use `expand_task` / `task-master expand --id=<id>` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc))
--   Otherwise use `expand_task` / `task-master expand --id=<id> --num=<number>`
--   Add `--research` flag to leverage Perplexity AI for research-backed expansion
--   Use `--prompt="<context>"` to provide additional context when needed
--   Review and adjust generated subtasks as necessary
--   Use `--all` flag with `expand` or `expand_all` to expand multiple pending tasks at once
--   If subtasks need regeneration, clear them first with `clear_subtasks` / `task-master clear-subtasks` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)).
+-   Use `expand_task` / `task-master expand --id=<id>`. It automatically uses the complexity report if found, otherwise generates default number of subtasks.
+-   Use `--num=<number>` to specify an explicit number of subtasks, overriding defaults or complexity report recommendations.
+-   Add `--research` flag to leverage Perplexity AI for research-backed expansion.
+-   Add `--force` flag to clear existing subtasks before generating new ones (default is to append).
+-   Use `--prompt="<context>"` to provide additional context when needed.
+-   Review and adjust generated subtasks as necessary.
+-   Use `expand_all` tool or `task-master expand --all` to expand multiple pending tasks at once, respecting flags like `--force` and `--research`.
+-   If subtasks need complete replacement (regardless of the `--force` flag on `expand`), clear them first with `clear_subtasks` / `task-master clear-subtasks --id=<id>`.
 
 ## Implementation Drift Handling
 
 -   When implementation differs significantly from planned approach
 -   When future tasks need modification due to current implementation choices
 -   When new dependencies or requirements emerge
--   Use `update` / `task-master update --from=<futureTaskId> --prompt='<explanation>\nUpdate context...'` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) to update multiple future tasks.
--   Use `update_task` / `task-master update-task --id=<taskId> --prompt='<explanation>\nUpdate context...'` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) to update a single specific task.
+-   Use `update` / `task-master update --from=<futureTaskId> --prompt='<explanation>\nUpdate context...' --research` to update multiple future tasks.
+-   Use `update_task` / `task-master update-task --id=<taskId> --prompt='<explanation>\nUpdate context...' --research` to update a single specific task.
 
 ## Task Status Management
 
@@ -97,28 +97,32 @@ Task Master offers two primary ways to interact:
 - **details**: In-depth implementation instructions (Example: `"Use GitHub client ID/secret, handle callback, set session token."`) 
 - **testStrategy**: Verification approach (Example: `"Deploy and call endpoint to confirm 'Hello World' response."`) 
 - **subtasks**: List of smaller, more specific tasks (Example: `[{"id": 1, "title": "Configure OAuth", ...}]`) 
-- Refer to [`tasks.mdc`](mdc:.cursor/rules/tasks.mdc) for more details on the task data structure.
+- Refer to task structure details (previously linked to `tasks.mdc`).
 
-## Environment Variables Configuration
+## Configuration Management (Updated)
 
-- Task Master behavior is configured via environment variables:
-  - **ANTHROPIC_API_KEY** (Required): Your Anthropic API key for Claude.
-  - **MODEL**: Claude model to use (e.g., `claude-3-opus-20240229`).
-  - **MAX_TOKENS**: Maximum tokens for AI responses.
-  - **TEMPERATURE**: Temperature for AI model responses.
-  - **DEBUG**: Enable debug logging (`true`/`false`).
-  - **LOG_LEVEL**: Console output level (`debug`, `info`, `warn`, `error`).
-  - **DEFAULT_SUBTASKS**: Default number of subtasks for `expand`.
-  - **DEFAULT_PRIORITY**: Default priority for new tasks.
-  - **PROJECT_NAME**: Project name used in metadata.
-  - **PROJECT_VERSION**: Project version used in metadata.
-  - **PERPLEXITY_API_KEY**: API key for Perplexity AI (for `--research` flags).
-  - **PERPLEXITY_MODEL**: Perplexity model to use (e.g., `sonar-medium-online`).
-- See [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc) for default values and examples.
+Taskmaster configuration is managed through two main mechanisms:
+
+1.  **`.taskmasterconfig` File (Primary):**
+    *   Located in the project root directory.
+    *   Stores most configuration settings: AI model selections (main, research, fallback), parameters (max tokens, temperature), logging level, default subtasks/priority, project name, etc.
+    *   **Managed via `task-master models --setup` command.** Do not edit manually unless you know what you are doing.
+    *   **View/Set specific models via `task-master models` command or `models` MCP tool.**
+    *   Created automatically when you run `task-master models --setup` for the first time.
+
+2.  **Environment Variables (`.env` / `mcp.json`):**
+    *   Used **only** for sensitive API keys and specific endpoint URLs.
+    *   Place API keys (one per provider) in a `.env` file in the project root for CLI usage.
+    *   For MCP/Cursor integration, configure these keys in the `env` section of `.cursor/mcp.json`.
+    *   Available keys/variables: See `assets/env.example` or the Configuration section in the command reference (previously linked to `taskmaster.mdc`).
+
+**Important:** Non-API key settings (like model selections, `MAX_TOKENS`, `LOG_LEVEL`) are **no longer configured via environment variables**. Use the `task-master models` command (or `--setup` for interactive configuration) or the `models` MCP tool.
+**If AI commands FAIL in MCP** verify that the API key for the selected provider is present in the `env` section of `.cursor/mcp.json`.
+**If AI commands FAIL in CLI** verify that the API key for the selected provider is present in the `.env` file in the root of the project.
 
 ## Determining the Next Task
 
-- Run `next_task` / `task-master next` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) to show the next task to work on
+- Run `next_task` / `task-master next` to show the next task to work on.
 - The command identifies tasks with all dependencies satisfied
 - Tasks are prioritized by priority level, dependency count, and ID
 - The command shows comprehensive task information including:
@@ -133,7 +137,7 @@ Task Master offers two primary ways to interact:
 
 ## Viewing Specific Task Details
 
-- Run `get_task` / `task-master show <id>` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) to view a specific task
+- Run `get_task` / `task-master show <id>` to view a specific task.
 - Use dot notation for subtasks: `task-master show 1.2` (shows subtask 2 of task 1)
 - Displays comprehensive information similar to the next command, but for a specific task
 - For parent tasks, shows all subtasks and their current status
@@ -143,8 +147,8 @@ Task Master offers two primary ways to interact:
 
 ## Managing Task Dependencies
 
-- Use `add_dependency` / `task-master add-dependency --id=<id> --depends-on=<id>` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) to add a dependency
-- Use `remove_dependency` / `task-master remove-dependency --id=<id> --depends-on=<id>` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)) to remove a dependency
+- Use `add_dependency` / `task-master add-dependency --id=<id> --depends-on=<id>` to add a dependency.
+- Use `remove_dependency` / `task-master remove-dependency --id=<id> --depends-on=<id>` to remove a dependency.
 - The system prevents circular dependencies and duplicate dependency entries
 - Dependencies are checked for existence before being added or removed
 - Task files are automatically regenerated after dependency changes
@@ -164,14 +168,14 @@ Once a task has been broken down into subtasks using `expand_task` or similar me
     *   Gather *all* relevant details from this exploration phase.
 
 3.  **Log the Plan:**
-    *   Run `update_subtask` / `task-master update-subtask --id=<subtaskId> --prompt='<detailed plan>'` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)).
+    *   Run `update_subtask` / `task-master update-subtask --id=<subtaskId> --prompt='<detailed plan>'`.
     *   Provide the *complete and detailed* findings from the exploration phase in the prompt. Include file paths, line numbers, proposed diffs, reasoning, and any potential challenges identified. Do not omit details. The goal is to create a rich, timestamped log within the subtask's `details`.
 
 4.  **Verify the Plan:**
     *   Run `get_task` / `task-master show <subtaskId>` again to confirm that the detailed implementation plan has been successfully appended to the subtask's details.
 
 5.  **Begin Implementation:**
-    *   Set the subtask status using `set_task_status` / `task-master set-status --id=<subtaskId> --status=in-progress` (see [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc)).
+    *   Set the subtask status using `set_task_status` / `task-master set-status --id=<subtaskId> --status=in-progress`.
     *   Start coding based on the logged plan.
 
 6.  **Refine and Log Progress (Iteration 2+):**
@@ -189,7 +193,7 @@ Once a task has been broken down into subtasks using `expand_task` or similar me
 7.  **Review & Update Rules (Post-Implementation):**
     *   Once the implementation for the subtask is functionally complete, review all code changes and the relevant chat history.
     *   Identify any new or modified code patterns, conventions, or best practices established during the implementation.
-    *   Create new or update existing Cursor rules in the `.cursor/rules/` directory to capture these patterns, following the guidelines in [`cursor_rules.mdc`](mdc:.cursor/rules/cursor_rules.mdc) and [`self_improve.mdc`](mdc:.cursor/rules/self_improve.mdc).
+    *   Create new or update existing rules following internal guidelines (previously linked to `cursor_rules.mdc` and `self_improve.mdc`).
 
 8.  **Mark Task Complete:**
     *   After verifying the implementation and updating any necessary rules, mark the subtask as completed: `set_task_status` / `task-master set-status --id=<subtaskId> --status=done`.
@@ -198,10 +202,10 @@ Once a task has been broken down into subtasks using `expand_task` or similar me
     *   Stage the relevant code changes and any updated/new rule files (`git add .`).
     *   Craft a comprehensive Git commit message summarizing the work done for the subtask, including both code implementation and any rule adjustments.
     *   Execute the commit command directly in the terminal (e.g., `git commit -m 'feat(module): Implement feature X for subtask <subtaskId>\n\n- Details about changes...\n- Updated rule Y for pattern Z'`).
-    *   Consider if a Changeset is needed according to [`changeset.mdc`](mdc:.cursor/rules/changeset.mdc). If so, run `npm run changeset`, stage the generated file, and amend the commit or create a new one.
+    *   Consider if a Changeset is needed according to internal versioning guidelines (previously linked to `changeset.mdc`). If so, run `npm run changeset`, stage the generated file, and amend the commit or create a new one.
 
 10. **Proceed to Next Subtask:**
-    *   Identify the next subtask in the dependency chain (e.g., using `next_task` / `task-master next`) and repeat this iterative process starting from step 1.
+    *   Identify the next subtask (e.g., using `next_task` / `task-master next`).
 
 ## Code Analysis & Refactoring Techniques
 

.cursor/rules/mcp.mdc

@@ -3,7 +3,6 @@ description: Guidelines for implementing and interacting with the Task Master MC
 globs: mcp-server/src/**/*, scripts/modules/**/*
 alwaysApply: false
 ---
-
 # Task Master MCP Server Guidelines
 
 This document outlines the architecture and implementation patterns for the Task Master Model Context Protocol (MCP) server, designed for integration with tools like Cursor.
@@ -90,69 +89,54 @@ When implementing a new direct function in `mcp-server/src/core/direct-functions
      ```
 
 5. **Handling Logging Context (`mcpLog`)**:
-   - **Requirement**: Core functions that use the internal `report` helper function (common in `task-manager.js`, `dependency-manager.js`, etc.) expect the `options` object to potentially contain an `mcpLog` property. This `mcpLog` object **must** have callable methods for each log level (e.g., `mcpLog.info(...)`, `mcpLog.error(...)`).
-   - **Challenge**: The `log` object provided by FastMCP to the direct function's context, while functional, might not perfectly match this expected structure or could change in the future. Passing it directly can lead to runtime errors like `mcpLog[level] is not a function`.
-   - **Solution: The Logger Wrapper Pattern**: To reliably bridge the FastMCP `log` object and the core function's `mcpLog` expectation, use a simple wrapper object within the direct function:
+   - **Requirement**: Core functions (like those in `task-manager.js`) may accept an `options` object containing an optional `mcpLog` property. If provided, the core function expects this object to have methods like `mcpLog.info(...)`, `mcpLog.error(...)`.
+   - **Solution: The Logger Wrapper Pattern**: When calling a core function from a direct function, pass the `log` object provided by FastMCP *wrapped* in the standard `logWrapper` object. This ensures the core function receives a logger with the expected method structure.
      ```javascript
      // Standard logWrapper pattern within a Direct Function
      const logWrapper = {
        info: (message, ...args) => log.info(message, ...args),
        warn: (message, ...args) => log.warn(message, ...args),
        error: (message, ...args) => log.error(message, ...args),
-       debug: (message, ...args) => log.debug && log.debug(message, ...args), // Handle optional debug
-       success: (message, ...args) => log.info(message, ...args) // Map success to info if needed
+       debug: (message, ...args) => log.debug && log.debug(message, ...args),
+       success: (message, ...args) => log.info(message, ...args)
      };
 
      // ... later when calling the core function ...
      await coreFunction(
        // ... other arguments ...
-       tasksPath, 
-       taskId,
        { 
          mcpLog: logWrapper, // Pass the wrapper object
-         session 
+         session // Also pass session if needed by core logic or AI service
        },
        'json' // Pass 'json' output format if supported by core function
      );
      ```
-   - **Critical For JSON Output Format**: Passing the `logWrapper` as `mcpLog` serves a dual purpose:
-     1. **Prevents Runtime Errors**: It ensures the `mcpLog[level](...)` calls within the core function succeed
-     2. **Controls Output Format**: In functions like `updateTaskById` and `updateSubtaskById`, the presence of `mcpLog` in the options triggers setting `outputFormat = 'json'` (instead of 'text'). This prevents UI elements (spinners, boxes) from being generated, which would break the JSON response.
-   - **Proven Solution**: This pattern has successfully fixed multiple issues in our MCP tools (including `update-task` and `update-subtask`), where direct passing of the `log` object or omitting `mcpLog` led to either runtime errors or JSON parsing failures from UI output.
-   - **When To Use**: Implement this wrapper in any direct function that calls a core function with an `options` object that might use `mcpLog` for logging or output format control.
-   - **Why it Works**: The `logWrapper` explicitly defines the `.info()`, `.warn()`, `.error()`, etc., methods that the core function's `report` helper needs, ensuring the `mcpLog[level](...)` call succeeds. It simply forwards the logging calls to the actual FastMCP `log` object.
-   - **Combined with Silent Mode**: Remember that using the `logWrapper` for `mcpLog` is **necessary *in addition* to using `enableSilentMode()` / `disableSilentMode()`** (see next point). The wrapper handles structured logging *within* the core function, while silent mode suppresses direct `console.log` and UI elements (spinners, boxes) that would break the MCP JSON response.
+   - **JSON Output**: Passing `mcpLog` (via the wrapper) often triggers the core function to use a JSON-friendly output format, suppressing spinners/boxes.
+   - ✅ **DO**: Implement this pattern in direct functions calling core functions that might use `mcpLog`.
 
 6. **Silent Mode Implementation**:
-    - ✅ **DO**: Import silent mode utilities at the top: `import { enableSilentMode, disableSilentMode, isSilentMode } from '../../../../scripts/modules/utils.js';`
-    - ✅ **DO**: Ensure core Task Master functions called from direct functions do **not** pollute `stdout` with console output (banners, spinners, logs) that would break MCP's JSON communication.
-      - **Preferred**: Modify the core function to accept an `outputFormat: 'json'` parameter and check it internally before printing UI elements. Pass `'json'` from the direct function.
-      - **Required Fallback/Guarantee**: If the core function cannot be modified or its output suppression is unreliable, **wrap the core function call** within the direct function using `enableSilentMode()` / `disableSilentMode()` in a `try/finally` block. This guarantees no console output interferes with the MCP response.
-    - ✅ **DO**: Use `isSilentMode()` function to check global silent mode status if needed (rare in direct functions), NEVER access the global `silentMode` variable directly.
-    - ❌ **DON'T**: Wrap AI client initialization or AI API calls in `enable/disableSilentMode`; their logging is controlled via the `log` object (passed potentially within the `logWrapper` for core functions).
-    - ❌ **DON'T**: Assume a core function is silent just because it *should* be. Verify or use the `enable/disableSilentMode` wrapper.
-    - **Example (Direct Function Guaranteeing Silence and using Log Wrapper)**:
+    - ✅ **DO**: Import silent mode utilities: `import { enableSilentMode, disableSilentMode, isSilentMode } from '../../../../scripts/modules/utils.js';`
+    - ✅ **DO**: Wrap core function calls *within direct functions* using `enableSilentMode()` / `disableSilentMode()` in a `try/finally` block if the core function might produce console output (spinners, boxes, direct `console.log`) that isn't reliably controlled by passing `{ mcpLog }` or an `outputFormat` parameter.
+    - ✅ **DO**: Always disable silent mode in the `finally` block.
+    - ❌ **DON'T**: Wrap calls to the unified AI service (`generateTextService`, `generateObjectService`) in silent mode; their logging is handled internally.
+    - **Example (Direct Function Guaranteeing Silence & using Log Wrapper)**:
       ```javascript
       export async function coreWrapperDirect(args, log, context = {}) {
         const { session } = context;
         const tasksPath = findTasksJsonPath(args, log);
-        
-        // Create the logger wrapper
-        const logWrapper = { /* ... as defined above ... */ };
+        const logWrapper = { /* ... */ };
 
         enableSilentMode(); // Ensure silence for direct console output
         try {
-          // Call core function, passing wrapper and 'json' format
           const result = await coreFunction(
              tasksPath,
              args.param1,
-             { mcpLog: logWrapper, session }, 
-             'json' // Explicitly request JSON format if supported
+             { mcpLog: logWrapper, session }, // Pass context
+             'json' // Request JSON format if supported
           );
           return { success: true, data: result };
         } catch (error) {
            log.error(`Error: ${error.message}`);
-           // Return standardized error object
            return { success: false, error: { /* ... */ } };
         } finally {
            disableSilentMode(); // Critical: Always disable in finally
@@ -163,32 +147,6 @@ When implementing a new direct function in `mcp-server/src/core/direct-functions
 7. **Debugging MCP/Core Logic Interaction**:
     - ✅ **DO**: If an MCP tool fails with unclear errors (like JSON parsing failures), run the equivalent `task-master` CLI command in the terminal. The CLI often provides more detailed error messages originating from the core logic (e.g., `ReferenceError`, stack traces) that are obscured by the MCP layer.
 
-### Specific Guidelines for AI-Based Direct Functions
-
-Direct functions that interact with AI (e.g., `addTaskDirect`, `expandTaskDirect`) have additional responsibilities:
-
-- **Context Parameter**: These functions receive an additional `context` object as their third parameter. **Critically, this object should only contain `{ session }`**. Do NOT expect or use `reportProgress` from this context.
-  ```javascript
-  export async function yourAIDirect(args, log, context = {}) {
-    const { session } = context; // Only expect session
-    // ...
-  }
-  ```
-- **AI Client Initialization**:
-  - ✅ **DO**: Use the utilities from [`mcp-server/src/core/utils/ai-client-utils.js`](mdc:mcp-server/src/core/utils/ai-client-utils.js) (e.g., `getAnthropicClientForMCP(session, log)`) to get AI client instances. These correctly use the `session` object to resolve API keys.
-  - ✅ **DO**: Wrap client initialization in a try/catch block and return a specific `AI_CLIENT_ERROR` on failure.
-- **AI Interaction**:
-  - ✅ **DO**: Build prompts using helper functions where appropriate (e.g., from `ai-prompt-helpers.js`).
-  - ✅ **DO**: Make the AI API call using appropriate helpers (e.g., `_handleAnthropicStream`). Pass the `log` object to these helpers for internal logging. **Do NOT pass `reportProgress`**.
-  - ✅ **DO**: Parse the AI response using helpers (e.g., `parseTaskJsonResponse`) and handle parsing errors with a specific code (e.g., `RESPONSE_PARSING_ERROR`).
-- **Calling Core Logic**:
-  - ✅ **DO**: After successful AI interaction, call the relevant core Task Master function (from `scripts/modules/`) if needed (e.g., `addTaskDirect` calls `addTask`).
-  - ✅ **DO**: Pass necessary data, including potentially the parsed AI results, to the core function.
-  - ✅ **DO**: If the core function can produce console output, call it with an `outputFormat: 'json'` argument (or similar, depending on the function) to suppress CLI output. Ensure the core function is updated to respect this. Use `enableSilentMode/disableSilentMode` around the core function call as a fallback if `outputFormat` is not supported or insufficient.
-- **Progress Indication**:
-  - ❌ **DON'T**: Call `reportProgress` within the direct function.
-  - ✅ **DO**: If intermediate progress status is needed *within* the long-running direct function, use standard logging: `log.info('Progress: Processing AI response...')`.
-
 ## Tool Definition and Execution
 
 ### Tool Structure
@@ -221,151 +179,78 @@ server.addTool({
 The `execute` function receives validated arguments and the FastMCP context:
 
 ```javascript
-// Standard signature
-execute: async (args, context) => {
-  // Tool implementation
-}
-
 // Destructured signature (recommended)
-execute: async (args, { log, reportProgress, session }) => {
+execute: async (args, { log, session }) => {
   // Tool implementation
 }
 ```
 
-- **args**: The first parameter contains all the validated parameters defined in the tool's schema.
-- **context**: The second parameter is an object containing `{ log, reportProgress, session }` provided by FastMCP.
-  - ✅ **DO**: Use `{ log, session }` when calling direct functions.
-  - ⚠️ **WARNING**: Avoid passing `reportProgress` down to direct functions due to client compatibility issues. See Progress Reporting Convention below.
+- **args**: Validated parameters.
+- **context**: Contains `{ log, session }` from FastMCP. (Removed `reportProgress`).
 
-### Standard Tool Execution Pattern
+### Standard Tool Execution Pattern with Path Normalization (Updated)
 
-The `execute` method within each MCP tool (in `mcp-server/src/tools/*.js`) should follow this standard pattern:
+To ensure consistent handling of project paths across different client environments (Windows, macOS, Linux, WSL) and input formats (e.g., `file:///...`, URI encoded paths), all MCP tool `execute` methods that require access to the project root **MUST** be wrapped with the `withNormalizedProjectRoot` Higher-Order Function (HOF).
 
-1.  **Log Entry**: Log the start of the tool execution with relevant arguments.
-2.  **Get Project Root**: Use the `getProjectRootFromSession(session, log)` utility (from [`tools/utils.js`](mdc:mcp-server/src/tools/utils.js)) to extract the project root path from the client session. Fall back to `args.projectRoot` if the session doesn't provide a root.
-3.  **Call Direct Function**: Invoke the corresponding `*Direct` function wrapper (e.g., `listTasksDirect` from [`task-master-core.js`](mdc:mcp-server/src/core/task-master-core.js)), passing an updated `args` object that includes the resolved `projectRoot`. Crucially, the third argument (context) passed to the direct function should **only include `{ log, session }`**. **Do NOT pass `reportProgress`**.
-    ```javascript
-    // Example call to a non-AI direct function
-    const result = await someDirectFunction({ ...args, projectRoot }, log); 
+This HOF, defined in [`mcp-server/src/tools/utils.js`](mdc:mcp-server/src/tools/utils.js), performs the following before calling the tool's core logic:
 
-    // Example call to an AI-based direct function
-    const resultAI = await someAIDirect({ ...args, projectRoot }, log, { session }); 
-    ```
-4.  **Handle Result**: Receive the result object (`{ success, data/error, fromCache }`) from the `*Direct` function.
-5.  **Format Response**: Pass this result object to the `handleApiResult` utility (from [`tools/utils.js`](mdc:mcp-server/src/tools/utils.js)) for standardized MCP response formatting and error handling.
-6.  **Return**: Return the formatted response object provided by `handleApiResult`.
+1.  **Determines the Raw Root:** It prioritizes `args.projectRoot` if provided by the client, otherwise it calls `getRawProjectRootFromSession` to extract the path from the session.
+2.  **Normalizes the Path:** It uses the `normalizeProjectRoot` helper to decode URIs, strip `file://` prefixes, fix potential Windows drive letter prefixes (e.g., `/C:/`), convert backslashes (`\`) to forward slashes (`/`), and resolve the path to an absolute path suitable for the server's OS.
+3.  **Injects Normalized Path:** It updates the `args` object by replacing the original `projectRoot` (or adding it) with the normalized, absolute path.
+4.  **Executes Original Logic:** It calls the original `execute` function body, passing the updated `args` object.
+
+**Implementation Example:**
 
 ```javascript
-// Example execute method structure for a tool calling an AI-based direct function
-import { getProjectRootFromSession, handleApiResult, createErrorResponse } from './utils.js';
-import { someAIDirectFunction } from '../core/task-master-core.js';
+// In mcp-server/src/tools/your-tool.js
+import {
+    handleApiResult,
+    createErrorResponse,
+    withNormalizedProjectRoot // <<< Import HOF
+} from './utils.js';
+import { yourDirectFunction } from '../core/task-master-core.js';
+import { findTasksJsonPath } from '../core/utils/path-utils.js'; // If needed
 
-// ... inside server.addTool({...})
-execute: async (args, { log, session }) => { // Note: reportProgress is omitted here
-  try {
-    log.info(`Starting AI tool execution with args: ${JSON.stringify(args)}`);
+export function registerYourTool(server) {
+    server.addTool({
+        name: "your_tool",
+        description: "...".
+        parameters: z.object({
+            // ... other parameters ...
+            projectRoot: z.string().optional().describe('...') // projectRoot is optional here, HOF handles fallback
+        }),
+        // Wrap the entire execute function
+        execute: withNormalizedProjectRoot(async (args, { log, session }) => {
+            // args.projectRoot is now guaranteed to be normalized and absolute
+            const { /* other args */, projectRoot } = args;
 
-    // 1. Get Project Root
-    let rootFolder = getProjectRootFromSession(session, log);
-    if (!rootFolder && args.projectRoot) { // Fallback if needed
-      rootFolder = args.projectRoot;
-      log.info(`Using project root from args as fallback: ${rootFolder}`);
-    }
+            try {
+                log.info(`Executing your_tool with normalized root: ${projectRoot}`);
 
-    // 2. Call AI-Based Direct Function (passing only log and session in context)
-    const result = await someAIDirectFunction({
-      ...args,
-      projectRoot: rootFolder // Ensure projectRoot is explicitly passed
-    }, log, { session }); // Pass session here, NO reportProgress
+                // Resolve paths using the normalized projectRoot
+                let tasksPath = findTasksJsonPath({ projectRoot, file: args.file }, log);
 
-    // 3. Handle and Format Response
-    return handleApiResult(result, log);
+                // Call direct function, passing normalized projectRoot if needed by direct func
+                const result = await yourDirectFunction(
+                    {
+                        /* other args */,
+                        projectRoot // Pass it if direct function needs it
+                    },
+                    log,
+                    { session }
+                );
 
-  } catch (error) {
-    log.error(`Error during AI tool execution: ${error.message}`);
-    return createErrorResponse(error.message);
-  }
+                return handleApiResult(result, log);
+            } catch (error) {
+                log.error(`Error in your_tool: ${error.message}`);
+                return createErrorResponse(error.message);
+            }
+        }) // End HOF wrap
+    });
 }
 ```
 
-### Using AsyncOperationManager for Background Tasks
-
-For tools that execute potentially long-running operations *where the AI call is just one part* (e.g., `expand-task`, `update`), use the AsyncOperationManager. The `add-task` command, as refactored, does *not* require this in the MCP tool layer because the direct function handles the primary AI work and returns the final result synchronously from the perspective of the MCP tool.
-
-For tools that *do* use `AsyncOperationManager`:
-
-```javascript
-import { AsyncOperationManager } from '../utils/async-operation-manager.js'; // Correct path assuming utils location
-import { getProjectRootFromSession, createContentResponse, createErrorResponse } from './utils.js';
-import { someIntensiveDirect } from '../core/task-master-core.js';
-
-// ... inside server.addTool({...})
-execute: async (args, { log, session }) => { // Note: reportProgress omitted
-  try {
-    log.info(`Starting background operation with args: ${JSON.stringify(args)}`);
-
-    // 1. Get Project Root
-    let rootFolder = getProjectRootFromSession(session, log);
-    if (!rootFolder && args.projectRoot) {
-      rootFolder = args.projectRoot;
-      log.info(`Using project root from args as fallback: ${rootFolder}`);
-    }
-    
-    // Create operation description
-    const operationDescription = `Expanding task ${args.id}...`; // Example
-
-    // 2. Start async operation using AsyncOperationManager
-    const operation = AsyncOperationManager.createOperation(
-      operationDescription,
-      async (reportProgressCallback) => { // This callback is provided by AsyncOperationManager
-        // This runs in the background
-        try {
-          // Report initial progress *from the manager's callback*
-          reportProgressCallback({ progress: 0, status: 'Starting operation...' });
-
-          // Call the direct function (passing only session context)
-          const result = await someIntensiveDirect(
-            { ...args, projectRoot: rootFolder }, 
-            log, 
-            { session } // Pass session, NO reportProgress
-          );
-
-          // Report final progress *from the manager's callback*
-          reportProgressCallback({
-            progress: 100,
-            status: result.success ? 'Operation completed' : 'Operation failed',
-            result: result.data, // Include final data if successful
-            error: result.error   // Include error object if failed
-          });
-
-          return result; // Return the direct function's result
-        } catch (error) {
-          // Handle errors within the async task
-           reportProgressCallback({
-            progress: 100,
-            status: 'Operation failed critically',
-            error: { message: error.message, code: error.code || 'ASYNC_OPERATION_FAILED' }
-          });
-          throw error; // Re-throw for the manager to catch
-        }
-      }
-    );
-
-    // 3. Return immediate response with operation ID
-    return {
-        status: 202, // StatusCodes.ACCEPTED
-        body: {
-            success: true,
-            message: 'Operation started',
-            operationId: operation.id
-        }
-    };
-  } catch (error) {
-    log.error(`Error starting background operation: ${error.message}`);
-    return createErrorResponse(`Failed to start operation: ${error.message}`); // Use standard error response
-  }
-}
-```
+By using this HOF, the core logic within the `execute` method and any downstream functions (like `findTasksJsonPath` or direct functions) can reliably expect `args.projectRoot` to be a clean, absolute path suitable for the server environment.
 
 ### Project Initialization Tool
 
@@ -417,19 +302,13 @@ log.error(`Error occurred: ${error.message}`, { stack: error.stack });
 log.info('Progress: 50% - AI call initiated...'); // Example progress logging
 ```
 
-### Progress Reporting Convention
-
-- ⚠️ **DEPRECATED within Direct Functions**: The `reportProgress` function passed in the `context` object should **NOT** be called from within `*Direct` functions. Doing so can cause client-side validation errors due to missing/incorrect `progressToken` handling.
-- ✅ **DO**: For tools using `AsyncOperationManager`, use the `reportProgressCallback` function *provided by the manager* within the background task definition (as shown in the `AsyncOperationManager` example above) to report progress updates for the *overall operation*.
-- ✅ **DO**: If finer-grained progress needs to be indicated *during* the execution of a `*Direct` function (whether called directly or via `AsyncOperationManager`), use `log.info()` statements (e.g., `log.info('Progress: Parsing AI response...')`).
-
-### Session Usage Convention
+## Session Usage Convention
 
 The `session` object (destructured from `context`) contains authenticated session data and client information.
 
 - **Authentication**: Access user-specific data (`session.userId`, etc.) if authentication is implemented.
 - **Project Root**: The primary use in Task Master is accessing `session.roots` to determine the client's project root directory via the `getProjectRootFromSession` utility (from [`tools/utils.js`](mdc:mcp-server/src/tools/utils.js)). See the Standard Tool Execution Pattern above.
-- **Environment Variables**: The `session.env` object is critical for AI tools. Pass the `session` object to the `*Direct` function's context, and then to AI client utility functions (like `getAnthropicClientForMCP`) which will extract API keys and other relevant environment settings (e.g., `MODEL`, `MAX_TOKENS`) from `session.env`.
+- **Environment Variables**: The `session.env` object provides access to environment variables set in the MCP client configuration (e.g., `.cursor/mcp.json`). This is the **primary mechanism** for the unified AI service layer (`ai-services-unified.js`) to securely access **API keys** when called from MCP context.
 - **Capabilities**: Can be used to check client capabilities (`session.clientCapabilities`).
 
 ## Direct Function Wrappers (`*Direct`)
@@ -438,24 +317,25 @@ These functions, located in `mcp-server/src/core/direct-functions/`, form the co
 
 - **Purpose**: Bridge MCP tools and core Task Master modules (`scripts/modules/*`). Handle AI interactions if applicable.
 - **Responsibilities**:
-  - Receive `args` (including the `projectRoot` determined by the tool), `log` object, and optionally a `context` object (containing **only `{ session }` if needed).
-  - **Find `tasks.json`**: Use `findTasksJsonPath(args, log)` from [`core/utils/path-utils.js`](mdc:mcp-server/src/core/utils/path-utils.js).
-  - Validate arguments specific to the core logic.
-  - **Handle AI Logic (if applicable)**: Initialize AI clients (using `session` from context), build prompts, make AI calls, parse responses.
-  - **Implement Caching (if applicable)**: Use `getCachedOrExecute` from [`tools/utils.js`](mdc:mcp-server/src/tools/utils.js) for read operations.
-  - **Call Core Logic**: Call the underlying function from the core Task Master modules, passing necessary data (including AI results if applicable).
-    - ✅ **DO**: Pass `outputFormat: 'json'` (or similar) to the core function if it might produce console output.
-    - ✅ **DO**: Wrap the core function call with `enableSilentMode/disableSilentMode` if necessary.
-  - Handle errors gracefully (AI errors, core logic errors, file errors).
-  - Return a standardized result object: `{ success: boolean, data?: any, error?: { code: string, message: string }, fromCache?: boolean }`.
-  - ❌ **DON'T**: Call `reportProgress`. Use `log.info` for progress indication if needed.
+  - Receive `args` (including `projectRoot`), `log`, and optionally `{ session }` context.
+  - Find `tasks.json` using `findTasksJsonPath`.
+  - Validate arguments.
+  - **Implement Caching (if applicable)**: Use `getCachedOrExecute`.
+  - **Call Core Logic**: Invoke function from `scripts/modules/*`.
+    - Pass `outputFormat: 'json'` if applicable.
+    - Wrap with `enableSilentMode/disableSilentMode` if needed.
+    - Pass `{ mcpLog: logWrapper, session }` context if core logic needs it.
+  - Handle errors.
+  - Return standardized result object.
+  - ❌ **DON'T**: Call `reportProgress`.
+  - ❌ **DON'T**: Initialize AI clients or call AI services directly.
 
 ## Key Principles
 
 - **Prefer Direct Function Calls**: MCP tools should always call `*Direct` wrappers instead of `executeTaskMasterCommand`.
 - **Standardized Execution Flow**: Follow the pattern: MCP Tool -> `getProjectRootFromSession` -> `*Direct` Function -> Core Logic / AI Logic.
 - **Path Resolution via Direct Functions**: The `*Direct` function is responsible for finding the exact `tasks.json` path using `findTasksJsonPath`, relying on the `projectRoot` passed in `args`.
-- **AI Logic in Direct Functions**: For AI-based tools, the `*Direct` function handles AI client initialization, calls, and parsing, using the `session` object passed in its context.
+- **AI Logic in Core Modules**: AI interactions (prompt building, calling unified service) reside within the core logic functions (`scripts/modules/*`), not direct functions.
 - **Silent Mode in Direct Functions**: Wrap *core function* calls (from `scripts/modules`) with `enableSilentMode()` and `disableSilentMode()` if they produce console output not handled by `outputFormat`. Do not wrap AI calls.
 - **Selective Async Processing**: Use `AsyncOperationManager` in the *MCP Tool layer* for operations involving multiple steps or long waits beyond a single AI call (e.g., file processing + AI call + file writing). Simple AI calls handled entirely within the `*Direct` function (like `addTaskDirect`) may not need it at the tool layer.
 - **No `reportProgress` in Direct Functions**: Do not pass or use `reportProgress` within `*Direct` functions. Use `log.info()` for internal progress or report progress from the `AsyncOperationManager` callback in the MCP tool layer.
@@ -480,7 +360,7 @@ Follow these steps to add MCP support for an existing Task Master command (see [
 
 1.  **Ensure Core Logic Exists**: Verify the core functionality is implemented and exported from the relevant module in `scripts/modules/`. Ensure the core function can suppress console output (e.g., via an `outputFormat` parameter).
 
-2.  **Create Direct Function File in `mcp-server/src/core/direct-functions/`**:
+2. **Create Direct Function File in `mcp-server/src/core/direct-functions/`**:
     - Create a new file (e.g., `your-command.js`) using **kebab-case** naming.
     - Import necessary core functions, `findTasksJsonPath`, silent mode utilities, and potentially AI client/prompt utilities.
     - Implement `async function yourCommandDirect(args, log, context = {})` using **camelCase** with `Direct` suffix. **Remember `context` should only contain `{ session }` if needed (for AI keys/config).**

.cursor/rules/new_features.mdc

@@ -25,11 +25,17 @@ alwaysApply: false
 The standard pattern for adding a feature follows this workflow:
 
 1. **Core Logic**: Implement the business logic in the appropriate module (e.g., [`task-manager.js`](mdc:scripts/modules/task-manager.js)).
-2. **UI Components**: Add any display functions to [`ui.js`](mdc:scripts/modules/ui.js) following [`ui.mdc`](mdc:.cursor/rules/ui.mdc).
-3. **Command Integration**: Add the CLI command to [`commands.js`](mdc:scripts/modules/commands.js) following [`commands.mdc`](mdc:.cursor/rules/commands.mdc).
-4. **Testing**: Write tests for all components of the feature (following [`tests.mdc`](mdc:.cursor/rules/tests.mdc))
-5. **Configuration**: Update any configuration in [`utils.js`](mdc:scripts/modules/utils.js) if needed, following [`utilities.mdc`](mdc:.cursor/rules/utilities.mdc).
-6. **Documentation**: Update help text and documentation in [dev_workflow.mdc](mdc:scripts/modules/dev_workflow.mdc)
+2. **AI Integration (If Applicable)**: 
+   - Import necessary service functions (e.g., `generateTextService`, `streamTextService`) from [`ai-services-unified.js`](mdc:scripts/modules/ai-services-unified.js).
+   - Prepare parameters (`role`, `session`, `systemPrompt`, `prompt`).
+   - Call the service function.
+   - Handle the response (direct text or stream object).
+   - **Important**: Prefer `generateTextService` for calls sending large context (like stringified JSON) where incremental display is not needed. See [`ai_services.mdc`](mdc:.cursor/rules/ai_services.mdc) for detailed usage patterns and cautions.
+3. **UI Components**: Add any display functions to [`ui.js`](mdc:scripts/modules/ui.js) following [`ui.mdc`](mdc:.cursor/rules/ui.mdc).
+4. **Command Integration**: Add the CLI command to [`commands.js`](mdc:scripts/modules/commands.js) following [`commands.mdc`](mdc:.cursor/rules/commands.mdc).
+5. **Testing**: Write tests for all components of the feature (following [`tests.mdc`](mdc:.cursor/rules/tests.mdc))
+6. **Configuration**: Update configuration settings or add new ones in [`config-manager.js`](mdc:scripts/modules/config-manager.js) and ensure getters/setters are appropriate. Update documentation in [`utilities.mdc`](mdc:.cursor/rules/utilities.mdc) and [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc). Update the `.taskmasterconfig` structure if needed.
+7. **Documentation**: Update help text and documentation in [`dev_workflow.mdc`](mdc:.cursor/rules/dev_workflow.mdc) and [`taskmaster.mdc`](mdc:.cursor/rules/taskmaster.mdc).
 
 ## Critical Checklist for New Features
 
@@ -211,7 +217,29 @@ export {
 ```
 
 ```javascript
-// 2. UI COMPONENTS: Add display function to ui.js
+// 2. AI Integration: Add import and use necessary service functions
+import { generateTextService } from './ai-services-unified.js';
+
+// Example usage:
+async function handleAIInteraction() {
+  const role = 'user';
+  const session = 'exampleSession';
+  const systemPrompt = 'You are a helpful assistant.';
+  const prompt = 'What is the capital of France?';
+
+  const result = await generateTextService(role, session, systemPrompt, prompt);
+  console.log(result);
+}
+
+// Export from the module
+export {
+  // ... existing exports ...
+  handleAIInteraction,
+};
+```
+
+```javascript
+// 3. UI COMPONENTS: Add display function to ui.js
 /**
  * Display archive operation results
  * @param {string} archivePath - Path to the archive file
@@ -232,7 +260,7 @@ export {
 ```
 
 ```javascript
-// 3. COMMAND INTEGRATION: Add to commands.js
+// 4. COMMAND INTEGRATION: Add to commands.js
 import { archiveTasks } from './task-manager.js';
 import { displayArchiveResults } from './ui.js';
 
@@ -452,7 +480,7 @@ npm test
 For each new feature:
 
 1. Add help text to the command definition
-2. Update [`dev_workflow.mdc`](mdc:scripts/modules/dev_workflow.mdc) with command reference
+2. Update [`dev_workflow.mdc`](mdc:.cursor/rules/dev_workflow.mdc) with command reference
 3. Consider updating [`architecture.mdc`](mdc:.cursor/rules/architecture.mdc) if the feature significantly changes module responsibilities.
 
 Follow the existing command reference format:
@@ -495,14 +523,24 @@ Integrating Task Master commands with the MCP server (for use by tools like Curs
 
 4.  **Create MCP Tool (`mcp-server/src/tools/`)**:
     - Create a new file (e.g., `your-command.js`) using **kebab-case**.
-    - Import `zod`, `handleApiResult`, `createErrorResponse`, **`getProjectRootFromSession`**, and your `yourCommandDirect` function.
+    - Import `zod`, `handleApiResult`, **`withNormalizedProjectRoot` HOF**, and your `yourCommandDirect` function.
     - Implement `registerYourCommandTool(server)`.
-    - Define the tool `name` using **snake_case** (e.g., `your_command`).
-    - Define the `parameters` using `zod`. **Crucially, define `projectRoot` as optional**: `projectRoot: z.string().optional().describe(...)`. Include `file` if applicable.
-    - Implement the standard `async execute(args, { log, reportProgress, session })` method:
-      - Get `rootFolder` using `getProjectRootFromSession` (with fallback to `args.projectRoot`).
-      - Call `yourCommandDirect({ ...args, projectRoot: rootFolder }, log)`. 
-      - Pass the result to `handleApiResult(result, log, 'Error Message')`.
+    - **Define parameters**: Make `projectRoot` optional (`z.string().optional().describe(...)`) as the HOF handles fallback.
+    - Consider if this operation should run in the background using `AsyncOperationManager`.
+    - Implement the standard `execute` method **wrapped with `withNormalizedProjectRoot`**:
+      ```javascript
+      execute: withNormalizedProjectRoot(async (args, { log, session }) => {
+          // args.projectRoot is now normalized
+          const { projectRoot /*, other args */ } = args;
+          // ... resolve tasks path if needed using normalized projectRoot ...
+          const result = await yourCommandDirect(
+              { /* other args */, projectRoot /* if needed by direct func */ }, 
+              log, 
+              { session }
+          );
+          return handleApiResult(result, log);
+      })
+      ```
 
 5.  **Register Tool**: Import and call `registerYourCommandTool` in `mcp-server/src/tools/index.js`.
 

.cursor/rules/self_improve.mdc

@@ -69,5 +69,4 @@ alwaysApply: true
   - Update references to external docs
   - Maintain links between related rules
   - Document breaking changes
-
 Follow [cursor_rules.mdc](mdc:.cursor/rules/cursor_rules.mdc) for proper rule formatting and structure.

.cursor/rules/taskmaster.mdc

@@ -3,14 +3,13 @@ description: Comprehensive reference for Taskmaster MCP tools and CLI commands.
 globs: **/*
 alwaysApply: true
 ---
-
 # Taskmaster Tool & Command Reference
 
-This document provides a detailed reference for interacting with Taskmaster, covering both the recommended MCP tools (for integrations like Cursor) and the corresponding `task-master` CLI commands (for direct user interaction or fallback).
+This document provides a detailed reference for interacting with Taskmaster, covering both the recommended MCP tools, suitable for integrations like Cursor, and the corresponding `task-master` CLI commands, designed for direct user interaction or fallback.
 
-**Note:** For interacting with Taskmaster programmatically or via integrated tools, using the **MCP tools is strongly recommended** due to better performance, structured data, and error handling. The CLI commands serve as a user-friendly alternative and fallback. See [`mcp.mdc`](mdc:.cursor/rules/mcp.mdc) for MCP implementation details and [`commands.mdc`](mdc:.cursor/rules/commands.mdc) for CLI implementation guidelines.
+**Note:** For interacting with Taskmaster programmatically or via integrated tools, using the **MCP tools is strongly recommended** due to better performance, structured data, and error handling. The CLI commands serve as a user-friendly alternative and fallback. 
 
-**Important:** Several MCP tools involve AI processing and are long-running operations that may take up to a minute to complete. When using these tools, always inform users that the operation is in progress and to wait patiently for results. The AI-powered tools include: `parse_prd`, `analyze_project_complexity`, `update_subtask`, `update_task`, `update`, `expand_all`, `expand_task`, and `add_task`.
+**Important:** Several MCP tools involve AI processing... The AI-powered tools include `parse_prd`, `analyze_project_complexity`, `update_subtask`, `update_task`, `update`, `expand_all`, `expand_task`, and `add_task`.
 
 ---
 
@@ -24,18 +23,18 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **Key CLI Options:**
     *   `--name <name>`: `Set the name for your project in Taskmaster's configuration.`
     *   `--description <text>`: `Provide a brief description for your project.`
-    *   `--version <version>`: `Set the initial version for your project (e.g., '0.1.0').`
+    *   `--version <version>`: `Set the initial version for your project, e.g., '0.1.0'.`
     *   `-y, --yes`: `Initialize Taskmaster quickly using default settings without interactive prompts.`
 *   **Usage:** Run this once at the beginning of a new project.
 *   **MCP Variant Description:** `Set up the basic Taskmaster file structure and configuration in the current directory for a new project by running the 'task-master init' command.`
 *   **Key MCP Parameters/Options:**
     *   `projectName`: `Set the name for your project.` (CLI: `--name <name>`)
     *   `projectDescription`: `Provide a brief description for your project.` (CLI: `--description <text>`)
-    *   `projectVersion`: `Set the initial version for your project (e.g., '0.1.0').` (CLI: `--version <version>`)
+    *   `projectVersion`: `Set the initial version for your project, e.g., '0.1.0'.` (CLI: `--version <version>`)
     *   `authorName`: `Author name.` (CLI: `--author <author>`)
-    *   `skipInstall`: `Skip installing dependencies (default: false).` (CLI: `--skip-install`)
-    *   `addAliases`: `Add shell aliases (tm, taskmaster) (default: false).` (CLI: `--aliases`)
-    *   `yes`: `Skip prompts and use defaults/provided arguments (default: false).` (CLI: `-y, --yes`)
+    *   `skipInstall`: `Skip installing dependencies. Default is false.` (CLI: `--skip-install`)
+    *   `addAliases`: `Add shell aliases tm and taskmaster. Default is false.` (CLI: `--aliases`)
+    *   `yes`: `Skip prompts and use defaults/provided arguments. Default is false.` (CLI: `-y, --yes`)
 *   **Usage:** Run this once at the beginning of a new project, typically via an integrated tool like Cursor. Operates on the current working directory of the MCP server. 
 *   **Important:** Once complete, you *MUST* parse a prd in order to generate tasks. There will be no tasks files until then. The next step after initializing should be to create a PRD using the example PRD in scripts/example_prd.txt. 
 
@@ -43,15 +42,45 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 
 *   **MCP Tool:** `parse_prd`
 *   **CLI Command:** `task-master parse-prd [file] [options]`
-*   **Description:** `Parse a Product Requirements Document (PRD) or text file with Taskmaster to automatically generate an initial set of tasks in tasks.json.`
+*   **Description:** `Parse a Product Requirements Document, PRD, or text file with Taskmaster to automatically generate an initial set of tasks in tasks.json.`
 *   **Key Parameters/Options:**
     *   `input`: `Path to your PRD or requirements text file that Taskmaster should parse for tasks.` (CLI: `[file]` positional or `-i, --input <file>`)
-    *   `output`: `Specify where Taskmaster should save the generated 'tasks.json' file (default: 'tasks/tasks.json').` (CLI: `-o, --output <file>`)
+    *   `output`: `Specify where Taskmaster should save the generated 'tasks.json' file. Defaults to 'tasks/tasks.json'.` (CLI: `-o, --output <file>`)
     *   `numTasks`: `Approximate number of top-level tasks Taskmaster should aim to generate from the document.` (CLI: `-n, --num-tasks <number>`)
     *   `force`: `Use this to allow Taskmaster to overwrite an existing 'tasks.json' without asking for confirmation.` (CLI: `-f, --force`)
 *   **Usage:** Useful for bootstrapping a project from an existing requirements document.
-*   **Notes:** Task Master will strictly adhere to any specific requirements mentioned in the PRD (libraries, database schemas, frameworks, tech stacks, etc.) while filling in any gaps where the PRD isn't fully specified. Tasks are designed to provide the most direct implementation path while avoiding over-engineering.
-*   **Important:** This MCP tool makes AI calls and can take up to a minute to complete. Please inform users to hang tight while the operation is in progress. If the user does not have a PRD, suggest discussing their idea and then use the example PRD in scripts/example_prd.txt as a template for creating the PRD based on their idea, for use with parse-prd.
+*   **Notes:** Task Master will strictly adhere to any specific requirements mentioned in the PRD, such as libraries, database schemas, frameworks, tech stacks, etc., while filling in any gaps where the PRD isn't fully specified. Tasks are designed to provide the most direct implementation path while avoiding over-engineering.
+*   **Important:** This MCP tool makes AI calls and can take up to a minute to complete. Please inform users to hang tight while the operation is in progress. If the user does not have a PRD, suggest discussing their idea and then use the example PRD in `scripts/example_prd.txt` as a template for creating the PRD based on their idea, for use with `parse-prd`.
+
+---
+
+## AI Model Configuration
+
+### 2. Manage Models (`models`)
+*   **MCP Tool:** `models`
+*   **CLI Command:** `task-master models [options]`
+*   **Description:** `View the current AI model configuration or set specific models for different roles (main, research, fallback). Allows setting custom model IDs for Ollama and OpenRouter.`
+*   **Key MCP Parameters/Options:**
+    *   `setMain <model_id>`: `Set the primary model ID for task generation/updates.` (CLI: `--set-main <model_id>`)
+    *   `setResearch <model_id>`: `Set the model ID for research-backed operations.` (CLI: `--set-research <model_id>`)
+    *   `setFallback <model_id>`: `Set the model ID to use if the primary fails.` (CLI: `--set-fallback <model_id>`)
+    *   `ollama <boolean>`: `Indicates the set model ID is a custom Ollama model.` (CLI: `--ollama`)
+    *   `openrouter <boolean>`: `Indicates the set model ID is a custom OpenRouter model.` (CLI: `--openrouter`)
+    *   `listAvailableModels <boolean>`: `If true, lists available models not currently assigned to a role.` (CLI: No direct equivalent; CLI lists available automatically)
+    *   `projectRoot <string>`: `Optional. Absolute path to the project root directory.` (CLI: Determined automatically)
+*   **Key CLI Options:**
+    *   `--set-main <model_id>`: `Set the primary model.`
+    *   `--set-research <model_id>`: `Set the research model.`
+    *   `--set-fallback <model_id>`: `Set the fallback model.`
+    *   `--ollama`: `Specify that the provided model ID is for Ollama (use with --set-*).`
+    *   `--openrouter`: `Specify that the provided model ID is for OpenRouter (use with --set-*). Validates against OpenRouter API.`
+    *   `--setup`: `Run interactive setup to configure models, including custom Ollama/OpenRouter IDs.`
+*   **Usage (MCP):** Call without set flags to get current config. Use `setMain`, `setResearch`, or `setFallback` with a valid model ID to update the configuration. Use `listAvailableModels: true` to get a list of unassigned models. To set a custom model, provide the model ID and set `ollama: true` or `openrouter: true`.
+*   **Usage (CLI):** Run without flags to view current configuration and available models. Use set flags to update specific roles. Use `--setup` for guided configuration, including custom models. To set a custom model via flags, use `--set-<role>=<model_id>` along with either `--ollama` or `--openrouter`.
+*   **Notes:** Configuration is stored in `.taskmasterconfig` in the project root. This command/tool modifies that file. Use `listAvailableModels` or `task-master models` to see internally supported models. OpenRouter custom models are validated against their live API. Ollama custom models are not validated live.
+*   **API note:** API keys for selected AI providers (based on their model) need to exist in the mcp.json file to be accessible in MCP context. The API keys must be present in the local .env file for the CLI to be able to read them.
+*   **Model costs:** The costs in supported models are expressed in dollars. An input/output value of 3 is $3.00. A value of 0.8 is $0.80. 
+*   **Warning:** DO NOT MANUALLY EDIT THE .taskmasterconfig FILE. Use the included commands either in the MCP or CLI format as needed. Always prioritize MCP tools when available and use the CLI as a fallback.
 
 ---
 
@@ -63,9 +92,9 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **CLI Command:** `task-master list [options]`
 *   **Description:** `List your Taskmaster tasks, optionally filtering by status and showing subtasks.`
 *   **Key Parameters/Options:**
-    *   `status`: `Show only Taskmaster tasks matching this status (e.g., 'pending', 'done').` (CLI: `-s, --status <status>`)
+    *   `status`: `Show only Taskmaster tasks matching this status, e.g., 'pending' or 'done'.` (CLI: `-s, --status <status>`)
     *   `withSubtasks`: `Include subtasks indented under their parent tasks in the list.` (CLI: `--with-subtasks`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Get an overview of the project status, often used at the start of a work session.
 
 ### 4. Get Next Task (`next_task`)
@@ -74,7 +103,7 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **CLI Command:** `task-master next [options]`
 *   **Description:** `Ask Taskmaster to show the next available task you can work on, based on status and completed dependencies.`
 *   **Key Parameters/Options:**
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Identify what to work on next according to the plan.
 
 ### 5. Get Task Details (`get_task`)
@@ -83,8 +112,8 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **CLI Command:** `task-master show [id] [options]`
 *   **Description:** `Display detailed information for a specific Taskmaster task or subtask by its ID.`
 *   **Key Parameters/Options:**
-    *   `id`: `Required. The ID of the Taskmaster task (e.g., '15') or subtask (e.g., '15.2') you want to view.` (CLI: `[id]` positional or `-i, --id <id>`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `id`: `Required. The ID of the Taskmaster task, e.g., '15', or subtask, e.g., '15.2', you want to view.` (CLI: `[id]` positional or `-i, --id <id>`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Understand the full details, implementation notes, and test strategy for a specific task before starting work.
 
 ---
@@ -97,10 +126,11 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **CLI Command:** `task-master add-task [options]`
 *   **Description:** `Add a new task to Taskmaster by describing it; AI will structure it.`
 *   **Key Parameters/Options:**
-    *   `prompt`: `Required. Describe the new task you want Taskmaster to create (e.g., "Implement user authentication using JWT").` (CLI: `-p, --prompt <text>`)
-    *   `dependencies`: `Specify the IDs of any Taskmaster tasks that must be completed before this new one can start (e.g., '12,14').` (CLI: `-d, --dependencies <ids>`)
-    *   `priority`: `Set the priority for the new task ('high', 'medium', 'low'; default: 'medium').` (CLI: `--priority <priority>`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `prompt`: `Required. Describe the new task you want Taskmaster to create, e.g., "Implement user authentication using JWT".` (CLI: `-p, --prompt <text>`)
+    *   `dependencies`: `Specify the IDs of any Taskmaster tasks that must be completed before this new one can start, e.g., '12,14'.` (CLI: `-d, --dependencies <ids>`)
+    *   `priority`: `Set the priority for the new task: 'high', 'medium', or 'low'. Default is 'medium'.` (CLI: `--priority <priority>`)
+    *   `research`: `Enable Taskmaster to use the research role for potentially more informed task creation.` (CLI: `-r, --research`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Quickly add newly identified tasks during development.
 *   **Important:** This MCP tool makes AI calls and can take up to a minute to complete. Please inform users to hang tight while the operation is in progress.
 
@@ -112,13 +142,13 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **Key Parameters/Options:**
     *   `id` / `parent`: `Required. The ID of the Taskmaster task that will be the parent.` (MCP: `id`, CLI: `-p, --parent <id>`)
     *   `taskId`: `Use this if you want to convert an existing top-level Taskmaster task into a subtask of the specified parent.` (CLI: `-i, --task-id <id>`)
-    *   `title`: `Required (if not using taskId). The title for the new subtask Taskmaster should create.` (CLI: `-t, --title <title>`)
+    *   `title`: `Required if not using taskId. The title for the new subtask Taskmaster should create.` (CLI: `-t, --title <title>`)
     *   `description`: `A brief description for the new subtask.` (CLI: `-d, --description <text>`)
     *   `details`: `Provide implementation notes or details for the new subtask.` (CLI: `--details <text>`)
-    *   `dependencies`: `Specify IDs of other tasks or subtasks (e.g., '15', '16.1') that must be done before this new subtask.` (CLI: `--dependencies <ids>`)
-    *   `status`: `Set the initial status for the new subtask (default: 'pending').` (CLI: `-s, --status <status>`)
+    *   `dependencies`: `Specify IDs of other tasks or subtasks, e.g., '15' or '16.1', that must be done before this new subtask.` (CLI: `--dependencies <ids>`)
+    *   `status`: `Set the initial status for the new subtask. Default is 'pending'.` (CLI: `-s, --status <status>`)
     *   `skipGenerate`: `Prevent Taskmaster from automatically regenerating markdown task files after adding the subtask.` (CLI: `--skip-generate`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Break down tasks manually or reorganize existing tasks.
 
 ### 8. Update Tasks (`update`)
@@ -127,10 +157,10 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **CLI Command:** `task-master update [options]`
 *   **Description:** `Update multiple upcoming tasks in Taskmaster based on new context or changes, starting from a specific task ID.`
 *   **Key Parameters/Options:**
-    *   `from`: `Required. The ID of the first task Taskmaster should update. All tasks with this ID or higher (and not 'done') will be considered.` (CLI: `--from <id>`)
-    *   `prompt`: `Required. Explain the change or new context for Taskmaster to apply to the tasks (e.g., "We are now using React Query instead of Redux Toolkit for data fetching").` (CLI: `-p, --prompt <text>`)
-    *   `research`: `Enable Taskmaster to use Perplexity AI for more informed updates based on external knowledge (requires PERPLEXITY_API_KEY).` (CLI: `-r, --research`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `from`: `Required. The ID of the first task Taskmaster should update. All tasks with this ID or higher that are not 'done' will be considered.` (CLI: `--from <id>`)
+    *   `prompt`: `Required. Explain the change or new context for Taskmaster to apply to the tasks, e.g., "We are now using React Query instead of Redux Toolkit for data fetching".` (CLI: `-p, --prompt <text>`)
+    *   `research`: `Enable Taskmaster to use the research role for more informed updates. Requires appropriate API key.` (CLI: `-r, --research`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Handle significant implementation changes or pivots that affect multiple future tasks. Example CLI: `task-master update --from='18' --prompt='Switching to React Query.\nNeed to refactor data fetching...'`
 *   **Important:** This MCP tool makes AI calls and can take up to a minute to complete. Please inform users to hang tight while the operation is in progress.
 
@@ -138,12 +168,12 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 
 *   **MCP Tool:** `update_task`
 *   **CLI Command:** `task-master update-task [options]`
-*   **Description:** `Modify a specific Taskmaster task (or subtask) by its ID, incorporating new information or changes.`
+*   **Description:** `Modify a specific Taskmaster task or subtask by its ID, incorporating new information or changes.`
 *   **Key Parameters/Options:**
-    *   `id`: `Required. The specific ID of the Taskmaster task (e.g., '15') or subtask (e.g., '15.2') you want to update.` (CLI: `-i, --id <id>`)
+    *   `id`: `Required. The specific ID of the Taskmaster task, e.g., '15', or subtask, e.g., '15.2', you want to update.` (CLI: `-i, --id <id>`)
     *   `prompt`: `Required. Explain the specific changes or provide the new information Taskmaster should incorporate into this task.` (CLI: `-p, --prompt <text>`)
-    *   `research`: `Enable Taskmaster to use Perplexity AI for more informed updates (requires PERPLEXITY_API_KEY).` (CLI: `-r, --research`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `research`: `Enable Taskmaster to use the research role for more informed updates. Requires appropriate API key.` (CLI: `-r, --research`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Refine a specific task based on new understanding or feedback. Example CLI: `task-master update-task --id='15' --prompt='Clarification: Use PostgreSQL instead of MySQL.\nUpdate schema details...'`
 *   **Important:** This MCP tool makes AI calls and can take up to a minute to complete. Please inform users to hang tight while the operation is in progress.
 
@@ -153,10 +183,10 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **CLI Command:** `task-master update-subtask [options]`
 *   **Description:** `Append timestamped notes or details to a specific Taskmaster subtask without overwriting existing content. Intended for iterative implementation logging.`
 *   **Key Parameters/Options:**
-    *   `id`: `Required. The specific ID of the Taskmaster subtask (e.g., '15.2') you want to add information to.` (CLI: `-i, --id <id>`)
+    *   `id`: `Required. The specific ID of the Taskmaster subtask, e.g., '15.2', you want to add information to.` (CLI: `-i, --id <id>`)
     *   `prompt`: `Required. Provide the information or notes Taskmaster should append to the subtask's details. Ensure this adds *new* information not already present.` (CLI: `-p, --prompt <text>`)
-    *   `research`: `Enable Taskmaster to use Perplexity AI for more informed updates (requires PERPLEXITY_API_KEY).` (CLI: `-r, --research`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `research`: `Enable Taskmaster to use the research role for more informed updates. Requires appropriate API key.` (CLI: `-r, --research`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Add implementation notes, code snippets, or clarifications to a subtask during development. Before calling, review the subtask's current details to append only fresh insights, helping to build a detailed log of the implementation journey and avoid redundancy. Example CLI: `task-master update-subtask --id='15.2' --prompt='Discovered that the API requires header X.\nImplementation needs adjustment...'`
 *   **Important:** This MCP tool makes AI calls and can take up to a minute to complete. Please inform users to hang tight while the operation is in progress.
 
@@ -164,11 +194,11 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 
 *   **MCP Tool:** `set_task_status`
 *   **CLI Command:** `task-master set-status [options]`
-*   **Description:** `Update the status of one or more Taskmaster tasks or subtasks (e.g., 'pending', 'in-progress', 'done').`
+*   **Description:** `Update the status of one or more Taskmaster tasks or subtasks, e.g., 'pending', 'in-progress', 'done'.`
 *   **Key Parameters/Options:**
-    *   `id`: `Required. The ID(s) of the Taskmaster task(s) or subtask(s) (e.g., '15', '15.2', '16,17.1') to update.` (CLI: `-i, --id <id>`)
-    *   `status`: `Required. The new status to set (e.g., 'done', 'pending', 'in-progress', 'review', 'cancelled').` (CLI: `-s, --status <status>`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `id`: `Required. The ID(s) of the Taskmaster task(s) or subtask(s), e.g., '15', '15.2', or '16,17.1', to update.` (CLI: `-i, --id <id>`)
+    *   `status`: `Required. The new status to set, e.g., 'done', 'pending', 'in-progress', 'review', 'cancelled'.` (CLI: `-s, --status <status>`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Mark progress as tasks move through the development cycle.
 
 ### 12. Remove Task (`remove_task`)
@@ -177,9 +207,9 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **CLI Command:** `task-master remove-task [options]`
 *   **Description:** `Permanently remove a task or subtask from the Taskmaster tasks list.`
 *   **Key Parameters/Options:**
-    *   `id`: `Required. The ID of the Taskmaster task (e.g., '5') or subtask (e.g., '5.2') to permanently remove.` (CLI: `-i, --id <id>`)
+    *   `id`: `Required. The ID of the Taskmaster task, e.g., '5', or subtask, e.g., '5.2', to permanently remove.` (CLI: `-i, --id <id>`)
     *   `yes`: `Skip the confirmation prompt and immediately delete the task.` (CLI: `-y, --yes`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Permanently delete tasks or subtasks that are no longer needed in the project.
 *   **Notes:** Use with caution as this operation cannot be undone. Consider using 'blocked', 'cancelled', or 'deferred' status instead if you just want to exclude a task from active planning but keep it for reference. The command automatically cleans up dependency references in other tasks.
 
@@ -191,28 +221,28 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 
 *   **MCP Tool:** `expand_task`
 *   **CLI Command:** `task-master expand [options]`
-*   **Description:** `Use Taskmaster's AI to break down a complex task (or all tasks) into smaller, manageable subtasks.`
+*   **Description:** `Use Taskmaster's AI to break down a complex task into smaller, manageable subtasks. Appends subtasks by default.`
 *   **Key Parameters/Options:**
     *   `id`: `The ID of the specific Taskmaster task you want to break down into subtasks.` (CLI: `-i, --id <id>`)
-    *   `num`: `Suggests how many subtasks Taskmaster should aim to create (uses complexity analysis by default).` (CLI: `-n, --num <number>`)
-    *   `research`: `Enable Taskmaster to use Perplexity AI for more informed subtask generation (requires PERPLEXITY_API_KEY).` (CLI: `-r, --research`)
-    *   `prompt`: `Provide extra context or specific instructions to Taskmaster for generating the subtasks.` (CLI: `-p, --prompt <text>`)
-    *   `force`: `Use this to make Taskmaster replace existing subtasks with newly generated ones.` (CLI: `--force`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
-*   **Usage:** Generate a detailed implementation plan for a complex task before starting coding.
+    *   `num`: `Optional: Suggests how many subtasks Taskmaster should aim to create. Uses complexity analysis/defaults otherwise.` (CLI: `-n, --num <number>`)
+    *   `research`: `Enable Taskmaster to use the research role for more informed subtask generation. Requires appropriate API key.` (CLI: `-r, --research`)
+    *   `prompt`: `Optional: Provide extra context or specific instructions to Taskmaster for generating the subtasks.` (CLI: `-p, --prompt <text>`)
+    *   `force`: `Optional: If true, clear existing subtasks before generating new ones. Default is false (append).` (CLI: `--force`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
+*   **Usage:** Generate a detailed implementation plan for a complex task before starting coding. Automatically uses complexity report recommendations if available and `num` is not specified.
 *   **Important:** This MCP tool makes AI calls and can take up to a minute to complete. Please inform users to hang tight while the operation is in progress.
 
 ### 14. Expand All Tasks (`expand_all`)
 
 *   **MCP Tool:** `expand_all`
 *   **CLI Command:** `task-master expand --all [options]` (Note: CLI uses the `expand` command with the `--all` flag)
-*   **Description:** `Tell Taskmaster to automatically expand all 'pending' tasks based on complexity analysis.`
+*   **Description:** `Tell Taskmaster to automatically expand all eligible pending/in-progress tasks based on complexity analysis or defaults. Appends subtasks by default.`
 *   **Key Parameters/Options:**
-    *   `num`: `Suggests how many subtasks Taskmaster should aim to create per task.` (CLI: `-n, --num <number>`)
-    *   `research`: `Enable Perplexity AI for more informed subtask generation (requires PERPLEXITY_API_KEY).` (CLI: `-r, --research`)
-    *   `prompt`: `Provide extra context for Taskmaster to apply generally during expansion.` (CLI: `-p, --prompt <text>`)
-    *   `force`: `Make Taskmaster replace existing subtasks.` (CLI: `--force`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `num`: `Optional: Suggests how many subtasks Taskmaster should aim to create per task.` (CLI: `-n, --num <number>`)
+    *   `research`: `Enable research role for more informed subtask generation. Requires appropriate API key.` (CLI: `-r, --research`)
+    *   `prompt`: `Optional: Provide extra context for Taskmaster to apply generally during expansion.` (CLI: `-p, --prompt <text>`)
+    *   `force`: `Optional: If true, clear existing subtasks before generating new ones for each eligible task. Default is false (append).` (CLI: `--force`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Useful after initial task generation or complexity analysis to break down multiple tasks at once.
 *   **Important:** This MCP tool makes AI calls and can take up to a minute to complete. Please inform users to hang tight while the operation is in progress.
 
@@ -222,9 +252,9 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **CLI Command:** `task-master clear-subtasks [options]`
 *   **Description:** `Remove all subtasks from one or more specified Taskmaster parent tasks.`
 *   **Key Parameters/Options:**
-    *   `id`: `The ID(s) of the Taskmaster parent task(s) whose subtasks you want to remove (e.g., '15', '16,18').` (Required unless using `all`) (CLI: `-i, --id <ids>`)
+    *   `id`: `The ID(s) of the Taskmaster parent task(s) whose subtasks you want to remove, e.g., '15' or '16,18'. Required unless using `all`.) (CLI: `-i, --id <ids>`)
     *   `all`: `Tell Taskmaster to remove subtasks from all parent tasks.` (CLI: `--all`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Used before regenerating subtasks with `expand_task` if the previous breakdown needs replacement.
 
 ### 16. Remove Subtask (`remove_subtask`)
@@ -233,10 +263,10 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **CLI Command:** `task-master remove-subtask [options]`
 *   **Description:** `Remove a subtask from its Taskmaster parent, optionally converting it into a standalone task.`
 *   **Key Parameters/Options:**
-    *   `id`: `Required. The ID(s) of the Taskmaster subtask(s) to remove (e.g., '15.2', '16.1,16.3').` (CLI: `-i, --id <id>`)
+    *   `id`: `Required. The ID(s) of the Taskmaster subtask(s) to remove, e.g., '15.2' or '16.1,16.3'.` (CLI: `-i, --id <id>`)
     *   `convert`: `If used, Taskmaster will turn the subtask into a regular top-level task instead of deleting it.` (CLI: `-c, --convert`)
     *   `skipGenerate`: `Prevent Taskmaster from automatically regenerating markdown task files after removing the subtask.` (CLI: `--skip-generate`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Delete unnecessary subtasks or promote a subtask to a top-level task.
 
 ---
@@ -250,8 +280,8 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **Description:** `Define a dependency in Taskmaster, making one task a prerequisite for another.`
 *   **Key Parameters/Options:**
     *   `id`: `Required. The ID of the Taskmaster task that will depend on another.` (CLI: `-i, --id <id>`)
-    *   `dependsOn`: `Required. The ID of the Taskmaster task that must be completed first (the prerequisite).` (CLI: `-d, --depends-on <id>`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `dependsOn`: `Required. The ID of the Taskmaster task that must be completed first, the prerequisite.` (CLI: `-d, --depends-on <id>`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <path>`)
 *   **Usage:** Establish the correct order of execution between tasks.
 
 ### 18. Remove Dependency (`remove_dependency`)
@@ -262,7 +292,7 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **Key Parameters/Options:**
     *   `id`: `Required. The ID of the Taskmaster task you want to remove a prerequisite from.` (CLI: `-i, --id <id>`)
     *   `dependsOn`: `Required. The ID of the Taskmaster task that should no longer be a prerequisite.` (CLI: `-d, --depends-on <id>`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Update task relationships when the order of execution changes.
 
 ### 19. Validate Dependencies (`validate_dependencies`)
@@ -271,7 +301,7 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **CLI Command:** `task-master validate-dependencies [options]`
 *   **Description:** `Check your Taskmaster tasks for dependency issues (like circular references or links to non-existent tasks) without making changes.`
 *   **Key Parameters/Options:**
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Audit the integrity of your task dependencies.
 
 ### 20. Fix Dependencies (`fix_dependencies`)
@@ -280,7 +310,7 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **CLI Command:** `task-master fix-dependencies [options]`
 *   **Description:** `Automatically fix dependency issues (like circular references or links to non-existent tasks) in your Taskmaster tasks.`
 *   **Key Parameters/Options:**
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Clean up dependency errors automatically.
 
 ---
@@ -295,8 +325,8 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **Key Parameters/Options:**
     *   `output`: `Where to save the complexity analysis report (default: 'scripts/task-complexity-report.json').` (CLI: `-o, --output <file>`)
     *   `threshold`: `The minimum complexity score (1-10) that should trigger a recommendation to expand a task.` (CLI: `-t, --threshold <number>`)
-    *   `research`: `Enable Perplexity AI for more accurate complexity analysis (requires PERPLEXITY_API_KEY).` (CLI: `-r, --research`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `research`: `Enable research role for more accurate complexity analysis. Requires appropriate API key.` (CLI: `-r, --research`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Used before breaking down tasks to identify which ones need the most attention.
 *   **Important:** This MCP tool makes AI calls and can take up to a minute to complete. Please inform users to hang tight while the operation is in progress.
 
@@ -320,34 +350,33 @@ This document provides a detailed reference for interacting with Taskmaster, cov
 *   **Description:** `Create or update individual Markdown files for each task based on your tasks.json.`
 *   **Key Parameters/Options:**
     *   `output`: `The directory where Taskmaster should save the task files (default: in a 'tasks' directory).` (CLI: `-o, --output <directory>`)
-    *   `file`: `Path to your Taskmaster 'tasks.json' file (default relies on auto-detection).` (CLI: `-f, --file <file>`)
+    *   `file`: `Path to your Taskmaster 'tasks.json' file. Default relies on auto-detection.` (CLI: `-f, --file <file>`)
 *   **Usage:** Run this after making changes to tasks.json to keep individual task files up to date.
 
 ---
 
-## Environment Variables Configuration
+## Environment Variables Configuration (Updated)
 
-Taskmaster's behavior can be customized via environment variables. These affect both CLI and MCP server operation:
+Taskmaster primarily uses the **`.taskmasterconfig`** file (in project root) for configuration (models, parameters, logging level, etc.), managed via `task-master models --setup`.
 
-*   **ANTHROPIC_API_KEY** (Required): Your Anthropic API key for Claude.
-*   **MODEL**: Claude model to use (default: `claude-3-opus-20240229`). 
-*   **MAX_TOKENS**: Maximum tokens for AI responses (default: 8192).
-*   **TEMPERATURE**: Temperature for AI model responses (default: 0.7).
-*   **DEBUG**: Enable debug logging (`true`/`false`, default: `false`).
-*   **LOG_LEVEL**: Console output level (`debug`, `info`, `warn`, `error`, default: `info`).
-*   **DEFAULT_SUBTASKS**: Default number of subtasks for `expand` (default: 5).
-*   **DEFAULT_PRIORITY**: Default priority for new tasks (default: `medium`).
-*   **PROJECT_NAME**: Project name used in metadata.
-*   **PROJECT_VERSION**: Project version used in metadata.
-*   **PERPLEXITY_API_KEY**: API key for Perplexity AI (for `--research` flags).
-*   **PERPLEXITY_MODEL**: Perplexity model to use (default: `sonar-medium-online`).
+Environment variables are used **only** for sensitive API keys related to AI providers and specific overrides like the Ollama base URL:
 
-Set these in your `.env` file in the project root or in your environment before running Taskmaster.
+*   **API Keys (Required for corresponding provider):**
+    *   `ANTHROPIC_API_KEY`
+    *   `PERPLEXITY_API_KEY`
+    *   `OPENAI_API_KEY`
+    *   `GOOGLE_API_KEY`
+    *   `MISTRAL_API_KEY`
+    *   `AZURE_OPENAI_API_KEY` (Requires `AZURE_OPENAI_ENDPOINT` too)
+    *   `OPENROUTER_API_KEY`
+    *   `XAI_API_KEY`
+    *   `OLLANA_API_KEY` (Requires `OLLAMA_BASE_URL` too)
+*   **Endpoints (Optional/Provider Specific inside .taskmasterconfig):**
+    *   `AZURE_OPENAI_ENDPOINT`
+    *   `OLLAMA_BASE_URL` (Default: `http://localhost:11434/api`)
+
+**Set API keys** in your **`.env`** file in the project root (for CLI use) or within the `env` section of your **`.cursor/mcp.json`** file (for MCP/Cursor integration). All other settings (model choice, max tokens, temperature, log level, custom endpoints) are managed in `.taskmasterconfig` via `task-master models` command or `models` MCP tool.
 
 ---
 
-For implementation details:
-*   CLI commands: See [`commands.mdc`](mdc:.cursor/rules/commands.mdc)
-*   MCP server: See [`mcp.mdc`](mdc:.cursor/rules/mcp.mdc)
-*   Task structure: See [`tasks.mdc`](mdc:.cursor/rules/tasks.mdc)
-*   Workflow: See [`dev_workflow.mdc`](mdc:.cursor/rules/dev_workflow.mdc)
+For details on how these commands fit into the development process, see the [Development Workflow Guide](mdc:.cursor/rules/dev_workflow.mdc).

.cursor/rules/tests.mdc

@@ -283,107 +283,97 @@ When testing ES modules (`"type": "module"` in package.json), traditional mockin
   - Imported functions may not use your mocked dependencies even with proper jest.mock() setup
   - ES module exports are read-only properties (cannot be reassigned during tests)
 
-- **Mocking Entire Modules**
-  ```javascript
-  // Mock the entire module with custom implementation
-  jest.mock('../../scripts/modules/task-manager.js', () => {
-    // Get original implementation for functions you want to preserve
-    const originalModule = jest.requireActual('../../scripts/modules/task-manager.js');
+- **Mocking Modules Statically Imported**
+  - For modules imported with standard `import` statements at the top level:
+    - Use `jest.mock('path/to/module', factory)` **before** any imports.
+    - Jest hoists these mocks.
+    - Ensure the factory function returns the mocked structure correctly.
 
-    // Return mix of original and mocked functionality
-    return {
-      ...originalModule,
-      generateTaskFiles: jest.fn() // Replace specific functions
-    };
+- **Mocking Dependencies for Dynamically Imported Modules**
+  - **Problem**: Standard `jest.mock()` often fails for dependencies of modules loaded later using dynamic `import('path/to/module')`. The mocks aren't applied correctly when the dynamic import resolves.
+  - **Solution**: Use `jest.unstable_mockModule(modulePath, factory)` **before** the dynamic `import()` call.
+  ```javascript
+  // 1. Define mock function instances
+  const mockExistsSync = jest.fn();
+  const mockReadFileSync = jest.fn();
+  // ... other mocks
+
+  // 2. Mock the dependency module *before* the dynamic import
+  jest.unstable_mockModule('fs', () => ({
+    __esModule: true, // Important for ES module mocks
+    // Mock named exports
+    existsSync: mockExistsSync,
+    readFileSync: mockReadFileSync,
+    // Mock default export if necessary
+    // default: { ... }
+  }));
+
+  // 3. Dynamically import the module under test (e.g., in beforeAll or test case)
+  let moduleUnderTest;
+  beforeAll(async () => {
+    // Ensure mocks are reset if needed before import
+    mockExistsSync.mockReset();
+    mockReadFileSync.mockReset();
+    // ... reset other mocks ...
+
+    // Import *after* unstable_mockModule is called
+    moduleUnderTest = await import('../../scripts/modules/module-using-fs.js');
   });
 
-  // Import after mocks
-  import * as taskManager from '../../scripts/modules/task-manager.js';
+  // 4. Now tests can use moduleUnderTest, and its 'fs' calls will hit the mocks
+  test('should use mocked fs.readFileSync', () => {
+    mockReadFileSync.mockReturnValue('mock data');
+    moduleUnderTest.readFileAndProcess();
+    expect(mockReadFileSync).toHaveBeenCalled();
+    // ... other assertions
+  });
+  ```
+  - ✅ **DO**: Call `jest.unstable_mockModule()` before `await import()`.
+  - ✅ **DO**: Include `__esModule: true` in the mock factory for ES modules.
+  - ✅ **DO**: Mock named and default exports as needed within the factory.
+  - ✅ **DO**: Reset mock functions (`mockFn.mockReset()`) before the dynamic import if they might have been called previously.
 
-  // Now you can use the mock directly
-  const { generateTaskFiles } = taskManager;
+- **Mocking Entire Modules (Static Import)**
+  ```javascript
+  // Mock the entire module with custom implementation for static imports
+  // ... (existing example remains valid) ...
   ```
 
 - **Direct Implementation Testing**
   - Instead of calling the actual function which may have module-scope reference issues:
   ```javascript
-  test('should perform expected actions', () => {
-    // Setup mocks for this specific test
-    mockReadJSON.mockImplementationOnce(() => sampleData);
-    
-    // Manually simulate the function's behavior
-    const data = mockReadJSON('path/file.json');
-    mockValidateAndFixDependencies(data, 'path/file.json');
-    
-    // Skip calling the actual function and verify mocks directly
-    expect(mockReadJSON).toHaveBeenCalledWith('path/file.json');
-    expect(mockValidateAndFixDependencies).toHaveBeenCalledWith(data, 'path/file.json');
-  });
+  // ... (existing example remains valid) ...
   ```
 
 - **Avoiding Module Property Assignment**
   ```javascript
-  // ❌ DON'T: This causes "Cannot assign to read only property" errors
-  const utils = await import('../../scripts/modules/utils.js');
-  utils.readJSON = mockReadJSON; // Error: read-only property
-  
-  // ✅ DO: Use the module factory pattern in jest.mock()
-  jest.mock('../../scripts/modules/utils.js', () => ({
-    readJSON: mockReadJSONFunc,
-    writeJSON: mockWriteJSONFunc
-  }));
+  // ... (existing example remains valid) ...
   ```
 
 - **Handling Mock Verification Failures**
   - If verification like `expect(mockFn).toHaveBeenCalled()` fails:
-    1. Check that your mock setup is before imports
-    2. Ensure you're using the right mock instance
-    3. Verify your test invokes behavior that would call the mock
-    4. Use `jest.clearAllMocks()` in beforeEach to reset mock state
-    5. Consider implementing a simpler test that directly verifies mock behavior
-
-- **Full Example Pattern**
-  ```javascript
-  // 1. Define mock implementations
-  const mockReadJSON = jest.fn();
-  const mockValidateAndFixDependencies = jest.fn();
-  
-  // 2. Mock modules
-  jest.mock('../../scripts/modules/utils.js', () => ({
-    readJSON: mockReadJSON,
-    // Include other functions as needed
-  }));
-  
-  jest.mock('../../scripts/modules/dependency-manager.js', () => ({
-    validateAndFixDependencies: mockValidateAndFixDependencies
-  }));
-  
-  // 3. Import after mocks
-  import * as taskManager from '../../scripts/modules/task-manager.js';
-  
-  describe('generateTaskFiles function', () => {
-    beforeEach(() => {
-      jest.clearAllMocks();
-    });
-    
-    test('should generate task files', () => {
-      // 4. Setup test-specific mock behavior
-      const sampleData = { tasks: [{ id: 1, title: 'Test' }] };
-      mockReadJSON.mockReturnValueOnce(sampleData);
-      
-      // 5. Create direct implementation test
-      // Instead of calling: taskManager.generateTaskFiles('path', 'dir')
-      
-      // Simulate reading data
-      const data = mockReadJSON('path');
-      expect(mockReadJSON).toHaveBeenCalledWith('path');
-      
-      // Simulate other operations the function would perform
-      mockValidateAndFixDependencies(data, 'path');
-      expect(mockValidateAndFixDependencies).toHaveBeenCalledWith(data, 'path');
-    });
-  });
-  ```
+    1.  Check that your mock setup (`jest.mock` or `jest.unstable_mockModule`) is correctly placed **before** imports (static or dynamic).
+    2.  Ensure you're using the right mock instance and it's properly passed to the module.
+    3.  Verify your test invokes behavior that *should* call the mock.
+    4.  Use `jest.clearAllMocks()` or specific `mockFn.mockReset()` in `beforeEach` to prevent state leakage between tests.
+    5.  **Check Console Assertions**: If verifying `console.log`, `console.warn`, or `console.error` calls, ensure your assertion matches the *actual* arguments passed. If the code logs a single formatted string, assert against that single string (using `expect.stringContaining` or exact match), not multiple `expect.stringContaining` arguments.
+        ```javascript
+        // Example: Code logs console.error(`Error: ${message}. Details: ${details}`)
+        // ❌ DON'T: Assert multiple arguments if only one is logged
+        // expect(console.error).toHaveBeenCalledWith(
+        //   expect.stringContaining('Error:'),
+        //   expect.stringContaining('Details:')
+        // );
+        // ✅ DO: Assert the single string argument
+        expect(console.error).toHaveBeenCalledWith(
+           expect.stringContaining('Error: Specific message. Details: More details')
+        );
+        // or for exact match:
+        expect(console.error).toHaveBeenCalledWith(
+           'Error: Specific message. Details: More details'
+        );
+        ```
+    6.  Consider implementing a simpler test that *only* verifies the mock behavior in isolation.
 
 ## Mocking Guidelines
 

.cursor/rules/utilities.mdc

@@ -3,7 +3,6 @@ description: Guidelines for implementing utility functions
 globs: scripts/modules/utils.js, mcp-server/src/**/*
 alwaysApply: false
 ---
-
 # Utility Function Guidelines
 
 ## General Principles
@@ -79,28 +78,30 @@ alwaysApply: false
   }
   ```
 
-## Configuration Management (in `scripts/modules/utils.js`)
+## Configuration Management (via `config-manager.js`)
 
-- **Environment Variables**:
-  - ✅ DO: Provide default values for all configuration
-  - ✅ DO: Use environment variables for customization
-  - ✅ DO: Document available configuration options
-  - ❌ DON'T: Hardcode values that should be configurable
+Taskmaster configuration (excluding API keys) is primarily managed through the `.taskmasterconfig` file located in the project root and accessed via getters in [`scripts/modules/config-manager.js`](mdc:scripts/modules/config-manager.js).
 
-  ```javascript
-  // ✅ DO: Set up configuration with defaults and environment overrides
-  const CONFIG = {
-    model: process.env.MODEL || 'claude-3-opus-20240229', // Updated default model
-    maxTokens: parseInt(process.env.MAX_TOKENS || '4000'),
-    temperature: parseFloat(process.env.TEMPERATURE || '0.7'),
-    debug: process.env.DEBUG === "true",
-    logLevel: process.env.LOG_LEVEL || "info",
-    defaultSubtasks: parseInt(process.env.DEFAULT_SUBTASKS || "3"),
-    defaultPriority: process.env.DEFAULT_PRIORITY || "medium",
-    projectName: process.env.PROJECT_NAME || "Task Master Project", // Generic project name
-    projectVersion: "1.5.0" // Version should be updated via release process
-  };
-  ```
+- **`.taskmasterconfig` File**:
+  - ✅ DO: Use this JSON file to store settings like AI model selections (main, research, fallback), parameters (temperature, maxTokens), logging level, default priority/subtasks, etc.
+  - ✅ DO: Manage this file using the `task-master models --setup` CLI command or the `models` MCP tool.
+  - ✅ DO: Rely on [`config-manager.js`](mdc:scripts/modules/config-manager.js) to load this file (using the correct project root passed from MCP or found via CLI utils), merge with defaults, and provide validated settings.
+  - ❌ DON'T: Store API keys in this file.
+  - ❌ DON'T: Manually edit this file unless necessary.
+
+- **Configuration Getters (`config-manager.js`)**:
+  - ✅ DO: Import and use specific getters from `config-manager.js` (e.g., `getMainProvider()`, `getLogLevel()`, `getMainMaxTokens()`) to access configuration values *needed for application logic* (like `getDefaultSubtasks`).
+  - ✅ DO: Pass the `explicitRoot` parameter to getters if calling from MCP direct functions to ensure the correct project's config is loaded.
+  - ❌ DON'T: Call AI-specific getters (like `getMainModelId`, `getMainMaxTokens`) from core logic functions (`scripts/modules/task-manager/*`). Instead, pass the `role` to the unified AI service.
+  - ❌ DON'T: Access configuration values directly from environment variables (except API keys).
+
+- **API Key Handling (`utils.js` & `ai-services-unified.js`)**:
+  - ✅ DO: Store API keys **only** in `.env` (for CLI, loaded by `dotenv` in `scripts/dev.js`) or `.cursor/mcp.json` (for MCP, accessed via `session.env`).
+  - ✅ DO: Use `isApiKeySet(providerName, session)` from `config-manager.js` to check if a provider's key is available *before* potentially attempting an AI call if needed, but note the unified service performs its own internal check.
+  - ✅ DO: Understand that the unified service layer (`ai-services-unified.js`) internally resolves API keys using `resolveEnvVariable(key, session)` from `utils.js`.
+
+- **Error Handling**:
+  - ✅ DO: Handle potential `ConfigurationError` if the `.taskmasterconfig` file is missing or invalid when accessed via `getConfig` (e.g., in `commands.js` or direct functions).
 
 ## Logging Utilities (in `scripts/modules/utils.js`)
 
@@ -427,36 +428,69 @@ alwaysApply: false
 
 ## MCP Server Tool Utilities (`mcp-server/src/tools/utils.js`)
 
-- **Purpose**: These utilities specifically support the MCP server tools ([`mcp-server/src/tools/*.js`](mdc:mcp-server/src/tools/*.js)), handling MCP communication patterns, response formatting, caching integration, and the CLI fallback mechanism.
-- **Refer to [`mcp.mdc`](mdc:.cursor/rules/mcp.mdc)** for detailed usage patterns within the MCP tool `execute` methods and direct function wrappers.
+These utilities specifically support the implementation and execution of MCP tools.
 
-- **`getProjectRootFromSession(session, log)`**: 
-    - ✅ **DO**: Call this utility **within the MCP tool's `execute` method** to extract the project root path from the `session` object.
-    - Decodes the `file://` URI and handles potential errors.
-    - Returns the project path string or `null`.
-    - The returned path should then be passed in the `args` object when calling the corresponding `*Direct` function (e.g., `yourDirectFunction({ ...args, projectRoot: rootFolder }, log)`).
+- **`normalizeProjectRoot(rawPath, log)`**:
+    - **Purpose**: Takes a raw project root path (potentially URI encoded, with `file://` prefix, Windows slashes) and returns a normalized, absolute path suitable for the server's OS.
+    - **Logic**: Decodes URI, strips `file://`, handles Windows drive prefix (`/C:/`), replaces `\` with `/`, uses `path.resolve()`.
+    - **Usage**: Used internally by `withNormalizedProjectRoot` HOF.
+
+- **`getRawProjectRootFromSession(session, log)`**:
+    - **Purpose**: Extracts the *raw* project root URI string from the session object (`session.roots[0].uri` or `session.roots.roots[0].uri`) without performing normalization.
+    - **Usage**: Used internally by `withNormalizedProjectRoot` HOF as a fallback if `args.projectRoot` isn't provided.
+
+- **`withNormalizedProjectRoot(executeFn)`**:
+    - **Purpose**: A Higher-Order Function (HOF) designed to wrap a tool's `execute` method.
+    - **Logic**: 
+        1. Determines the raw project root (from `args.projectRoot` or `getRawProjectRootFromSession`).
+        2. Normalizes the raw path using `normalizeProjectRoot`.
+        3. Injects the normalized, absolute path back into the `args` object as `args.projectRoot`.
+        4. Calls the original `executeFn` with the updated `args`.
+    - **Usage**: Should wrap the `execute` function of *every* MCP tool that needs a reliable, normalized project root path.
+    - **Example**:
+      ```javascript
+      // In mcp-server/src/tools/your-tool.js
+      import { withNormalizedProjectRoot } from './utils.js';
+      
+      export function registerYourTool(server) {
+          server.addTool({
+              // ... name, description, parameters ...
+              execute: withNormalizedProjectRoot(async (args, context) => {
+                  // args.projectRoot is now normalized here
+                  const { projectRoot /*, other args */ } = args;
+                  // ... rest of tool logic using normalized projectRoot ...
+              })
+          });
+      }
+      ```
 
 - **`handleApiResult(result, log, errorPrefix, processFunction)`**:
-    - ✅ **DO**: Call this from the MCP tool's `execute` method after receiving the result from the `*Direct` function wrapper.
-    - Takes the standard `{ success, data/error, fromCache }` object.
-    - Formats the standard MCP success or error response, including the `fromCache` flag.
-    - Uses `processMCPResponseData` by default to filter response data.
-
-- **`executeTaskMasterCommand(command, log, args, projectRootRaw)`**:
-    - Executes a Task Master CLI command as a child process.
-    - Handles fallback between global `task-master` and local `node scripts/dev.js`.
-    - ❌ **DON'T**: Use this as the primary method for MCP tools. Prefer direct function calls via `*Direct` wrappers.
-
-- **`processMCPResponseData(taskOrData, fieldsToRemove)`**:
-    - Filters task data (e.g., removing `details`, `testStrategy`) before sending to the MCP client. Called by `handleApiResult`.
+    - **Purpose**: Standardizes the formatting of responses returned by direct functions (`{ success, data/error, fromCache }`) into the MCP response format.
+    - **Usage**: Call this at the end of the tool's `execute` method, passing the result from the direct function call.
 
 - **`createContentResponse(content)` / `createErrorResponse(errorMessage)`**:
-    - Formatters for standard MCP success/error responses.
+    - **Purpose**: Helper functions to create the basic MCP response structure for success or error messages.
+    - **Usage**: Used internally by `handleApiResult` and potentially directly for simple responses.
+
+- **`createLogWrapper(log)`**:
+    - **Purpose**: Creates a logger object wrapper with standard methods (`info`, `warn`, `error`, `debug`, `success`) mapping to the passed MCP `log` object's methods. Ensures compatibility when passing loggers to core functions.
+    - **Usage**: Used within direct functions before passing the `log` object down to core logic that expects the standard method names.
 
 - **`getCachedOrExecute({ cacheKey, actionFn, log })`**:
-    - ✅ **DO**: Use this utility *inside direct function wrappers* to implement caching.
-    - Checks cache, executes `actionFn` on miss, stores result.
-    - Returns standard `{ success, data/error, fromCache: boolean }`.
+    - **Purpose**: Utility for implementing caching within direct functions. Checks cache for `cacheKey`; if miss, executes `actionFn`, caches successful result, and returns.
+    - **Usage**: Wrap the core logic execution within a direct function call.
+
+- **`processMCPResponseData(taskOrData, fieldsToRemove)`**:
+    - **Purpose**: Utility to filter potentially sensitive or large fields (like `details`, `testStrategy`) from task objects before sending the response back via MCP.
+    - **Usage**: Passed as the default `processFunction` to `handleApiResult`.
+
+- **`getProjectRootFromSession(session, log)`**:
+    - **Purpose**: Legacy function to extract *and normalize* the project root from the session. Replaced by the HOF pattern but potentially still used.
+    - **Recommendation**: Prefer using the `withNormalizedProjectRoot` HOF in tools instead of calling this directly.
+
+- **`executeTaskMasterCommand(...)`**: 
+    - **Purpose**: Executes `task-master` CLI command as a fallback. 
+    - **Recommendation**: Deprecated for most uses; prefer direct function calls.
 
 ## Export Organization
 

.env.example

@@ -1,20 +1,9 @@
-# API Keys (Required)
-ANTHROPIC_API_KEY=your_anthropic_api_key_here       # Format: sk-ant-api03-...
-PERPLEXITY_API_KEY=your_perplexity_api_key_here     # Format: pplx-...
-
-# Model Configuration
-MODEL=claude-3-7-sonnet-20250219                    # Recommended models: claude-3-7-sonnet-20250219, claude-3-opus-20240229
-PERPLEXITY_MODEL=sonar-pro                          # Perplexity model for research-backed subtasks
-MAX_TOKENS=64000                                    # Maximum tokens for model responses
-TEMPERATURE=0.2                                     # Temperature for model responses (0.0-1.0)
-
-# Logging Configuration
-DEBUG=false                                         # Enable debug logging (true/false)
-LOG_LEVEL=info                                      # Log level (debug, info, warn, error)
-
-# Task Generation Settings
-DEFAULT_SUBTASKS=5                                  # Default number of subtasks when expanding
-DEFAULT_PRIORITY=medium                             # Default priority for generated tasks (high, medium, low)
-
-# Project Metadata (Optional)
-    PROJECT_NAME=Your Project Name                  # Override default project name in tasks.json
+# API Keys (Required for using in any role i.e. main/research/fallback -- see `task-master models`)
+ANTHROPIC_API_KEY=YOUR_ANTHROPIC_KEY_HERE
+PERPLEXITY_API_KEY=YOUR_PERPLEXITY_KEY_HERE
+OPENAI_API_KEY=YOUR_OPENAI_KEY_HERE
+GOOGLE_API_KEY=YOUR_GOOGLE_KEY_HERE
+MISTRAL_API_KEY=YOUR_MISTRAL_KEY_HERE
+OPENROUTER_API_KEY=YOUR_OPENROUTER_KEY_HERE
+XAI_API_KEY=YOUR_XAI_KEY_HERE
+AZURE_OPENAI_API_KEY=YOUR_AZURE_KEY_HERE

.github/workflows/release.yml

@@ -3,6 +3,9 @@ on:
   push:
     branches:
       - main
+
+concurrency: ${{ github.workflow }}-${{ github.ref }}
+
 jobs:
   release:
     runs-on: ubuntu-latest

.gitignore

@@ -19,6 +19,8 @@ npm-debug.log*
 yarn-debug.log*
 yarn-error.log*
 lerna-debug.log*
+tests/e2e/_runs/
+tests/e2e/log/
 
 # Coverage directory used by tools like istanbul
 coverage
@@ -59,3 +61,6 @@ dist
 *.debug
 init-debug.log
 dev-debug.log
+
+# NPMRC
+.npmrc

.taskmasterconfig

@@ -0,0 +1,31 @@
+{
+  "models": {
+    "main": {
+      "provider": "anthropic",
+      "modelId": "claude-3-7-sonnet-20250219",
+      "maxTokens": 100000,
+      "temperature": 0.2
+    },
+    "research": {
+      "provider": "perplexity",
+      "modelId": "sonar-pro",
+      "maxTokens": 8700,
+      "temperature": 0.1
+    },
+    "fallback": {
+      "provider": "anthropic",
+      "modelId": "claude-3-7-sonnet-20250219",
+      "maxTokens": 120000,
+      "temperature": 0.2
+    }
+  },
+  "global": {
+    "logLevel": "info",
+    "debug": false,
+    "defaultSubtasks": 5,
+    "defaultPriority": "medium",
+    "projectName": "Taskmaster",
+    "ollamaBaseUrl": "http://localhost:11434/api",
+    "azureOpenaiBaseUrl": "https://your-endpoint.openai.azure.com/"
+  }
+}

CHANGELOG.md

@@ -1,5 +1,100 @@
 # task-master-ai
 
+## 0.13.0-rc.1
+
+### Patch Changes
+
+- 4cbfe82: Fix ERR_MODULE_NOT_FOUND when trying to run MCP Server
+
+## 0.13.0-rc.0
+
+### Minor Changes
+
+- ef782ff: feat(expand): Enhance `expand` and `expand-all` commands
+
+  - Integrate `task-complexity-report.json` to automatically determine the number of subtasks and use tailored prompts for expansion based on prior analysis. You no longer need to try copy-pasting the recommended prompt. If it exists, it will use it for you. You can just run `task-master update --id=[id of task] --research` and it will use that prompt automatically. No extra prompt needed.
+  - Change default behavior to _append_ new subtasks to existing ones. Use the `--force` flag to clear existing subtasks before expanding. This is helpful if you need to add more subtasks to a task but you want to do it by the batch from a given prompt. Use force if you want to start fresh with a task's subtasks.
+
+- 87d97bb: Adds support for the OpenRouter AI provider. Users can now configure models available through OpenRouter (requiring an `OPENROUTER_API_KEY`) via the `task-master models` command, granting access to a wide range of additional LLMs. - IMPORTANT FYI ABOUT OPENROUTER: Taskmaster relies on AI SDK, which itself relies on tool use. It looks like **free** models sometimes do not include tool use. For example, Gemini 2.5 pro (free) failed via OpenRouter (no tool use) but worked fine on the paid version of the model. Custom model support for Open Router is considered experimental and likely will not be further improved for some time.
+- 1ab836f: Adds model management and new configuration file .taskmasterconfig which houses the models used for main, research and fallback. Adds models command and setter flags. Adds a --setup flag with an interactive setup. We should be calling this during init. Shows a table of active and available models when models is called without flags. Includes SWE scores and token costs, which are manually entered into the supported_models.json, the new place where models are defined for support. Config-manager.js is the core module responsible for managing the new config."
+- c8722b0: Adds custom model ID support for Ollama and OpenRouter providers.
+  - Adds the `--ollama` and `--openrouter` flags to `task-master models --set-<role>` command to set models for those providers outside of the support models list.
+  - Updated `task-master models --setup` interactive mode with options to explicitly enter custom Ollama or OpenRouter model IDs.
+  - Implemented live validation against OpenRouter API (`/api/v1/models`) when setting a custom OpenRouter model ID (via flag or setup).
+  - Refined logic to prioritize explicit provider flags/choices over internal model list lookups in case of ID conflicts.
+  - Added warnings when setting custom/unvalidated models.
+  - We obviously don't recommend going with a custom, unproven model. If you do and find performance is good, please let us know so we can add it to the list of supported models.
+- 2517bc1: Integrate OpenAI as a new AI provider. - Enhance `models` command/tool to display API key status. - Implement model-specific `maxTokens` override based on `supported-models.json` to save you if you use an incorrect max token value.
+- 9a48278: Tweaks Perplexity AI calls for research mode to max out input tokens and get day-fresh information - Forces temp at 0.1 for highly deterministic output, no variations - Adds a system prompt to further improve the output - Correctly uses the maximum input tokens (8,719, used 8,700) for perplexity - Specificies to use a high degree of research across the web - Specifies to use information that is as fresh as today; this support stuff like capturing brand new announcements like new GPT models and being able to query for those in research. 🔥
+
+### Patch Changes
+
+- 842eaf7: - Add support for Google Gemini models via Vercel AI SDK integration.
+- ed79d4f: Add xAI provider and Grok models support
+- ad89253: Better support for file paths on Windows, Linux & WSL.
+
+  - Standardizes handling of different path formats (URI encoded, Windows, Linux, WSL).
+  - Ensures tools receive a clean, absolute path suitable for the server OS.
+  - Simplifies tool implementation by centralizing normalization logic.
+
+- 2acba94: Add integration for Roo Code
+- d63964a: Improved update-subtask - Now it has context about the parent task details - It also has context about the subtask before it and the subtask after it (if they exist) - Not passing all subtasks to stay token efficient
+- 5f504fa: Improve and adjust `init` command for robustness and updated dependencies.
+
+  - **Update Initialization Dependencies:** Ensure newly initialized projects (`task-master init`) include all required AI SDK dependencies (`@ai-sdk/*`, `ai`, provider wrappers) in their `package.json` for out-of-the-box AI feature compatibility. Remove unnecessary dependencies (e.g., `uuid`) from the init template.
+  - **Silence `npm install` during `init`:** Prevent `npm install` output from interfering with non-interactive/MCP initialization by suppressing its stdio in silent mode.
+  - **Improve Conditional Model Setup:** Reliably skip interactive `models --setup` during non-interactive `init` runs (e.g., `init -y` or MCP) by checking `isSilentMode()` instead of passing flags.
+  - **Refactor `init.js`:** Remove internal `isInteractive` flag logic.
+  - **Update `init` Instructions:** Tweak the "Getting Started" text displayed after `init`.
+  - **Fix MCP Server Launch:** Update `.cursor/mcp.json` template to use `node ./mcp-server/server.js` instead of `npx task-master-mcp`.
+  - **Update Default Model:** Change the default main model in the `.taskmasterconfig` template.
+
+- 96aeeff: Fixes an issue with add-task which did not use the manually defined properties and still needlessly hit the AI endpoint.
+- 5aea93d: Fixes an issue that prevented remove-subtask with comma separated tasks/subtasks from being deleted (only the first ID was being deleted). Closes #140
+- 66ac9ab: Improves next command to be subtask-aware - The logic for determining the "next task" (findNextTask function, used by task-master next and the next_task MCP tool) has been significantly improved. Previously, it only considered top-level tasks, making its recommendation less useful when a parent task containing subtasks was already marked 'in-progress'. - The updated logic now prioritizes finding the next available subtask within any 'in-progress' parent task, considering subtask dependencies and priority. - If no suitable subtask is found within active parent tasks, it falls back to recommending the next eligible top-level task based on the original criteria (status, dependencies, priority).
+
+  This change makes the next command much more relevant and helpful during the implementation phase of complex tasks.
+
+- ca7b045: Add `--status` flag to `show` command to filter displayed subtasks.
+- 5a2371b: Fix --task to --num-tasks in ui + related tests - issue #324
+- 6cb213e: Adds a 'models' CLI and MCP command to get the current model configuration, available models, and gives the ability to set main/research/fallback models." - In the CLI, `task-master models` shows the current models config. Using the `--setup` flag launches an interactive set up that allows you to easily select the models you want to use for each of the three roles. Use `q` during the interactive setup to cancel the setup. - In the MCP, responses are simplified in RESTful format (instead of the full CLI output). The agent can use the `models` tool with different arguments, including `listAvailableModels` to get available models. Run without arguments, it will return the current configuration. Arguments are available to set the model for each of the three roles. This allows you to manage Taskmaster AI providers and models directly from either the CLI or MCP or both. - Updated the CLI help menu when you run `task-master` to include missing commands and .taskmasterconfig information. - Adds `--research` flag to `add-task` so you can hit up Perplexity right from the add-task flow, rather than having to add a task and then update it.
+
+## 0.12.1
+
+### Patch Changes
+
+- [#307](https://github.com/eyaltoledano/claude-task-master/pull/307) [`2829194`](https://github.com/eyaltoledano/claude-task-master/commit/2829194d3c1dd5373d3bf40275cf4f63b12d49a7) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Fix add_dependency tool crashing the MCP Server
+
+## 0.12.0
+
+### Minor Changes
+
+- [#253](https://github.com/eyaltoledano/claude-task-master/pull/253) [`b2ccd60`](https://github.com/eyaltoledano/claude-task-master/commit/b2ccd605264e47a61451b4c012030ee29011bb40) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Add `npx task-master-ai` that runs mcp instead of using `task-master-mcp``
+
+- [#267](https://github.com/eyaltoledano/claude-task-master/pull/267) [`c17d912`](https://github.com/eyaltoledano/claude-task-master/commit/c17d912237e6caaa2445e934fc48cd4841abf056) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Improve PRD parsing prompt with structured analysis and clearer task generation guidelines. We are testing a new prompt - please provide feedback on your experience.
+
+### Patch Changes
+
+- [#243](https://github.com/eyaltoledano/claude-task-master/pull/243) [`454a1d9`](https://github.com/eyaltoledano/claude-task-master/commit/454a1d9d37439c702656eedc0702c2f7a4451517) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - - Fixes shebang issue not allowing task-master to run on certain windows operating systems
+
+  - Resolves #241 #211 #184 #193
+
+- [#268](https://github.com/eyaltoledano/claude-task-master/pull/268) [`3e872f8`](https://github.com/eyaltoledano/claude-task-master/commit/3e872f8afbb46cd3978f3852b858c233450b9f33) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Fix remove-task command to handle multiple comma-separated task IDs
+
+- [#239](https://github.com/eyaltoledano/claude-task-master/pull/239) [`6599cb0`](https://github.com/eyaltoledano/claude-task-master/commit/6599cb0bf9eccecab528207836e9d45b8536e5c2) Thanks [@eyaltoledano](https://github.com/eyaltoledano)! - Updates the parameter descriptions for update, update-task and update-subtask to ensure the MCP server correctly reaches for the right update command based on what is being updated -- all tasks, one task, or a subtask.
+
+- [#272](https://github.com/eyaltoledano/claude-task-master/pull/272) [`3aee9bc`](https://github.com/eyaltoledano/claude-task-master/commit/3aee9bc840eb8f31230bd1b761ed156b261cabc4) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Enhance the `parsePRD` to include `--append` flag. This flag allows users to append the parsed PRD to an existing file, making it easier to manage multiple PRD files without overwriting existing content.
+
+- [#264](https://github.com/eyaltoledano/claude-task-master/pull/264) [`ff8e75c`](https://github.com/eyaltoledano/claude-task-master/commit/ff8e75cded91fb677903040002626f7a82fd5f88) Thanks [@joedanz](https://github.com/joedanz)! - Add quotes around numeric env vars in mcp.json (Windsurf, etc.)
+
+- [#248](https://github.com/eyaltoledano/claude-task-master/pull/248) [`d99fa00`](https://github.com/eyaltoledano/claude-task-master/commit/d99fa00980fc61695195949b33dcda7781006f90) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - - Fix `task-master init` polluting codebase with new packages inside `package.json` and modifying project `README`
+
+  - Now only initializes with cursor rules, windsurf rules, mcp.json, scripts/example_prd.txt, .gitignore modifications, and `README-task-master.md`
+
+- [#266](https://github.com/eyaltoledano/claude-task-master/pull/266) [`41b979c`](https://github.com/eyaltoledano/claude-task-master/commit/41b979c23963483e54331015a86e7c5079f657e4) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Fixed a bug that prevented the task-master from running in a Linux container
+
+- [#265](https://github.com/eyaltoledano/claude-task-master/pull/265) [`0eb16d5`](https://github.com/eyaltoledano/claude-task-master/commit/0eb16d5ecbb8402d1318ca9509e9d4087b27fb25) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Remove the need for project name, description, and version. Since we no longer create a package.json for you
+
 ## 0.11.0
 
 ### Minor Changes

README-task-master.md

@@ -13,25 +13,22 @@ A task management system for AI-driven development with Claude, designed to work
 
 ## Configuration
 
-The script can be configured through environment variables in a `.env` file at the root of the project:
+Taskmaster uses two primary configuration methods:
 
-### Required Configuration
+1.  **`.taskmasterconfig` File (Project Root)**
 
-- `ANTHROPIC_API_KEY`: Your Anthropic API key for Claude
+    - Stores most settings: AI model selections (main, research, fallback), parameters (max tokens, temperature), logging level, default priority/subtasks, project name.
+    - **Created and managed using `task-master models --setup` CLI command or the `models` MCP tool.**
+    - Do not edit manually unless you know what you are doing.
 
-### Optional Configuration
+2.  **Environment Variables (`.env` file or MCP `env` block)**
+    - Used **only** for sensitive **API Keys** (e.g., `ANTHROPIC_API_KEY`, `PERPLEXITY_API_KEY`, etc.) and specific endpoints (like `OLLAMA_BASE_URL`).
+    - **For CLI:** Place keys in a `.env` file in your project root.
+    - **For MCP/Cursor:** Place keys in the `env` section of your `.cursor/mcp.json` (or other MCP config according to the AI IDE or client you use) file under the `taskmaster-ai` server definition.
 
-- `MODEL`: Specify which Claude model to use (default: "claude-3-7-sonnet-20250219")
-- `MAX_TOKENS`: Maximum tokens for model responses (default: 4000)
-- `TEMPERATURE`: Temperature for model responses (default: 0.7)
-- `PERPLEXITY_API_KEY`: Your Perplexity API key for research-backed subtask generation
-- `PERPLEXITY_MODEL`: Specify which Perplexity model to use (default: "sonar-medium-online")
-- `DEBUG`: Enable debug logging (default: false)
-- `LOG_LEVEL`: Log level - debug, info, warn, error (default: info)
-- `DEFAULT_SUBTASKS`: Default number of subtasks when expanding (default: 3)
-- `DEFAULT_PRIORITY`: Default priority for generated tasks (default: medium)
-- `PROJECT_NAME`: Override default project name in tasks.json
-- `PROJECT_VERSION`: Override default version in tasks.json
+**Important:** Settings like model choices, max tokens, temperature, and log level are **no longer configured via environment variables.** Use the `task-master models` command or tool.
+
+See the [Configuration Guide](docs/configuration.md) for full details.
 
 ## Installation
 
@@ -50,7 +47,7 @@ npm install task-master-ai
 task-master init
 
 # If installed locally
-npx task-master-init
+npx task-master init
 ```
 
 This will prompt you for project details and set up a new project with the necessary files and structure.

README.md

@@ -1,6 +1,6 @@
 # Task Master [![GitHub stars](https://img.shields.io/github/stars/eyaltoledano/claude-task-master?style=social)](https://github.com/eyaltoledano/claude-task-master/stargazers)
 
-[![CI](https://github.com/eyaltoledano/claude-task-master/actions/workflows/ci.yml/badge.svg)](https://github.com/eyaltoledano/claude-task-master/actions/workflows/ci.yml) [![npm version](https://badge.fury.io/js/task-master-ai.svg)](https://badge.fury.io/js/task-master-ai) ![Discord Follow](https://dcbadge.limes.pink/api/server/https://discord.gg/2ms58QJjqp?style=flat) [![License: MIT with Commons Clause](https://img.shields.io/badge/license-MIT%20with%20Commons%20Clause-blue.svg)](LICENSE)
+[![CI](https://github.com/eyaltoledano/claude-task-master/actions/workflows/ci.yml/badge.svg)](https://github.com/eyaltoledano/claude-task-master/actions/workflows/ci.yml) [![npm version](https://badge.fury.io/js/task-master-ai.svg)](https://badge.fury.io/js/task-master-ai) [![Discord](https://dcbadge.limes.pink/api/server/https://discord.gg/taskmasterai?style=flat)](https://discord.gg/taskmasterai) [![License: MIT with Commons Clause](https://img.shields.io/badge/license-MIT%20with%20Commons%20Clause-blue.svg)](LICENSE)
 
 ### By [@eyaltoledano](https://x.com/eyaltoledano) & [@RalphEcom](https://x.com/RalphEcom)
 
@@ -27,16 +27,16 @@ MCP (Model Control Protocol) provides the easiest way to get started with Task M
 	"mcpServers": {
 		"taskmaster-ai": {
 			"command": "npx",
-			"args": ["-y", "task-master-ai"],
+			"args": ["-y", "--package=task-master-ai", "task-master-ai"],
 			"env": {
 				"ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
 				"PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY_HERE",
-				"MODEL": "claude-3-7-sonnet-20250219",
-				"PERPLEXITY_MODEL": "sonar-pro",
-				"MAX_TOKENS": 64000,
-				"TEMPERATURE": 0.2,
-				"DEFAULT_SUBTASKS": 5,
-				"DEFAULT_PRIORITY": "medium"
+				"OPENAI_API_KEY": "YOUR_OPENAI_KEY_HERE",
+				"GOOGLE_API_KEY": "YOUR_GOOGLE_KEY_HERE",
+				"MISTRAL_API_KEY": "YOUR_MISTRAL_KEY_HERE",
+				"OPENROUTER_API_KEY": "YOUR_OPENROUTER_KEY_HERE",
+				"XAI_API_KEY": "YOUR_XAI_KEY_HERE",
+				"AZURE_OPENAI_API_KEY": "YOUR_AZURE_KEY_HERE"
 			}
 		}
 	}
@@ -79,7 +79,7 @@ npm install task-master-ai
 task-master init
 
 # If installed locally
-npx task-master-init
+npx task-master init
 ```
 
 This will prompt you for project details and set up a new project with the necessary files and structure.

assets/.taskmasterconfig

@@ -0,0 +1,31 @@
+{
+	"models": {
+		"main": {
+			"provider": "anthropic",
+			"modelId": "claude-3-7-sonnet-20250219",
+			"maxTokens": 120000,
+			"temperature": 0.2
+		},
+		"research": {
+			"provider": "perplexity",
+			"modelId": "sonar-pro",
+			"maxTokens": 8700,
+			"temperature": 0.1
+		},
+		"fallback": {
+			"provider": "anthropic",
+			"modelId": "claude-3.5-sonnet-20240620",
+			"maxTokens": 120000,
+			"temperature": 0.1
+		}
+	},
+	"global": {
+		"logLevel": "info",
+		"debug": false,
+		"defaultSubtasks": 5,
+		"defaultPriority": "medium",
+		"projectName": "Taskmaster",
+		"ollamaBaseUrl": "http://localhost:11434/api",
+		"azureOpenaiBaseUrl": "https://your-endpoint.openai.azure.com/"
+	}
+}

assets/env.example

@@ -1,14 +1,8 @@
-# Required
-ANTHROPIC_API_KEY=your-api-key-here  # For most AI ops -- Format: sk-ant-api03-... (Required)
-PERPLEXITY_API_KEY=pplx-abcde        # For research -- Format: pplx-abcde (Optional, Highly Recommended)
-
-# Optional - defaults shown
-MODEL=claude-3-7-sonnet-20250219  # Recommended models: claude-3-7-sonnet-20250219, claude-3-opus-20240229 (Required)
-PERPLEXITY_MODEL=sonar-pro        # Make sure you have access to sonar-pro otherwise you can use sonar regular (Optional)
-MAX_TOKENS=64000                   # Maximum tokens for model responses (Required)
-TEMPERATURE=0.2                   # Temperature for model responses (0.0-1.0) - lower = less creativity and follow your prompt closely (Required)
-DEBUG=false                       # Enable debug logging (true/false)
-LOG_LEVEL=info                    # Log level (debug, info, warn, error)
-DEFAULT_SUBTASKS=5                # Default number of subtasks when expanding
-DEFAULT_PRIORITY=medium           # Default priority for generated tasks (high, medium, low)
-PROJECT_NAME={{projectName}}      # Project name for tasks.json metadata
+# API Keys (Required to enable respective provider)
+ANTHROPIC_API_KEY=your_anthropic_api_key_here       # Required: Format: sk-ant-api03-...
+PERPLEXITY_API_KEY=your_perplexity_api_key_here     # Optional: Format: pplx-...
+OPENAI_API_KEY=your_openai_api_key_here             # Optional, for OpenAI/OpenRouter models. Format: sk-proj-...
+GOOGLE_API_KEY=your_google_api_key_here             # Optional, for Google Gemini models.
+MISTRAL_API_KEY=your_mistral_key_here               # Optional, for Mistral AI models.
+XAI_API_KEY=YOUR_XAI_KEY_HERE                       # Optional, for xAI AI models.
+AZURE_OPENAI_API_KEY=your_azure_key_here            # Optional, for Azure OpenAI models (requires endpoint in .taskmasterconfig).

assets/roocode/.roo/rules-architect/architect-rules

@@ -0,0 +1,93 @@
+**Core Directives & Agentivity:**
+# 1. Adhere strictly to the rules defined below.
+# 2. Use tools sequentially, one per message. Adhere strictly to the rules defined below.
+# 3. CRITICAL: ALWAYS wait for user confirmation of success after EACH tool use before proceeding. Do not assume success.
+# 4. Operate iteratively: Analyze task -> Plan steps -> Execute steps one by one.
+# 5. Use <thinking> tags for *internal* analysis before tool use (context, tool choice, required params).
+# 6. **DO NOT DISPLAY XML TOOL TAGS IN THE OUTPUT.**
+# 7. **DO NOT DISPLAY YOUR THINKING IN THE OUTPUT.**
+
+**Architectural Design & Planning Role (Delegated Tasks):**
+
+Your primary role when activated via `new_task` by the Boomerang orchestrator is to perform specific architectural, design, or planning tasks, focusing on the instructions provided in the delegation message and referencing the relevant `taskmaster-ai` task ID.
+
+1.  **Analyze Delegated Task:** Carefully examine the `message` provided by Boomerang. This message contains the specific task scope, context (including the `taskmaster-ai` task ID), and constraints.
+2.  **Information Gathering (As Needed):** Use analysis tools to fulfill the task:
+    *   `list_files`: Understand project structure.
+    *   `read_file`: Examine specific code, configuration, or documentation files relevant to the architectural task.
+    *   `list_code_definition_names`: Analyze code structure and relationships.
+    *   `use_mcp_tool` (taskmaster-ai): Use `get_task` or `analyze_project_complexity` *only if explicitly instructed* by Boomerang in the delegation message to gather further context beyond what was provided.
+3.  **Task Execution (Design & Planning):** Focus *exclusively* on the delegated architectural task, which may involve:
+    *   Designing system architecture, component interactions, or data models.
+    *   Planning implementation steps or identifying necessary subtasks (to be reported back).
+    *   Analyzing technical feasibility, complexity, or potential risks.
+    *   Defining interfaces, APIs, or data contracts.
+    *   Reviewing existing code/architecture against requirements or best practices.
+4.  **Reporting Completion:** Signal completion using `attempt_completion`. Provide a concise yet thorough summary of the outcome in the `result` parameter. This summary is **crucial** for Boomerang to update `taskmaster-ai`. Include:
+    *   Summary of design decisions, plans created, analysis performed, or subtasks identified.
+    *   Any relevant artifacts produced (e.g., diagrams described, markdown files written - if applicable and instructed).
+    *   Completion status (success, failure, needs review).
+    *   Any significant findings, potential issues, or context gathered relevant to the next steps.
+5.  **Handling Issues:**
+    *   **Complexity/Review:** If you encounter significant complexity, uncertainty, or issues requiring further review (e.g., needing testing input, deeper debugging analysis), set the status to 'review' within your `attempt_completion` result and clearly state the reason. **Do not delegate directly.** Report back to Boomerang.
+    *   **Failure:** If the task fails (e.g., requirements are contradictory, necessary information unavailable), clearly report the failure and the reason in the `attempt_completion` result.
+6.  **Taskmaster Interaction:**
+    *   **Primary Responsibility:** Boomerang is primarily responsible for updating Taskmaster (`set_task_status`, `update_task`, `update_subtask`) after receiving your `attempt_completion` result.
+    *   **Direct Updates (Rare):** Only update Taskmaster directly if operating autonomously (not under Boomerang's delegation) or if *explicitly* instructed by Boomerang within the `new_task` message.
+7.  **Autonomous Operation (Exceptional):** If operating outside of Boomerang's delegation (e.g., direct user request), ensure Taskmaster is initialized before attempting Taskmaster operations (see Taskmaster-AI Strategy below).
+
+**Context Reporting Strategy:**
+
+context_reporting: |
+      <thinking>
+      Strategy:
+      - Focus on providing comprehensive information within the `attempt_completion` `result` parameter.
+      - Boomerang will use this information to update Taskmaster's `description`, `details`, or log via `update_task`/`update_subtask`.
+      - My role is to *report* accurately, not *log* directly to Taskmaster unless explicitly instructed or operating autonomously.
+      </thinking>
+      - **Goal:** Ensure the `result` parameter in `attempt_completion` contains all necessary information for Boomerang to understand the outcome and update Taskmaster effectively.
+      - **Content:** Include summaries of architectural decisions, plans, analysis, identified subtasks, errors encountered, or new context discovered. Structure the `result` clearly.
+      - **Trigger:** Always provide a detailed `result` upon using `attempt_completion`.
+      - **Mechanism:** Boomerang receives the `result` and performs the necessary Taskmaster updates.
+
+**Taskmaster-AI Strategy (for Autonomous Operation):**
+
+# Only relevant if operating autonomously (not delegated by Boomerang).
+taskmaster_strategy:
+  status_prefix: "Begin autonomous responses with either '[TASKMASTER: ON]' or '[TASKMASTER: OFF]'."
+  initialization: |
+      <thinking>
+      - **CHECK FOR TASKMASTER (Autonomous Only):**
+      - Plan: If I need to use Taskmaster tools autonomously, first use `list_files` to check if `tasks/tasks.json` exists.
+      - If `tasks/tasks.json` is present = set TASKMASTER: ON, else TASKMASTER: OFF.
+      </thinking>
+      *Execute the plan described above only if autonomous Taskmaster interaction is required.*
+  if_uninitialized: |
+      1. **Inform:** "Task Master is not initialized. Autonomous Taskmaster operations cannot proceed."
+      2. **Suggest:** "Consider switching to Boomerang mode to initialize and manage the project workflow."
+  if_ready: |
+      1. **Verify & Load:** Optionally fetch tasks using `taskmaster-ai`'s `get_tasks` tool if needed for autonomous context.
+      2. **Set Status:** Set status to '[TASKMASTER: ON]'.
+      3. **Proceed:** Proceed with autonomous Taskmaster operations.
+
+**Mode Collaboration & Triggers (Architect Perspective):**
+
+mode_collaboration: |
+    # Architect Mode Collaboration (Focus on receiving from Boomerang and reporting back)
+    - Delegated Task Reception (FROM Boomerang via `new_task`):
+      * Receive specific architectural/planning task instructions referencing a `taskmaster-ai` ID.
+      * Analyze requirements, scope, and constraints provided by Boomerang.
+    - Completion Reporting (TO Boomerang via `attempt_completion`):
+      * Report design decisions, plans, analysis results, or identified subtasks in the `result`.
+      * Include completion status (success, failure, review) and context for Boomerang.
+      * Signal completion of the *specific delegated architectural task*.
+
+mode_triggers:
+  # Conditions that might trigger a switch TO Architect mode (typically orchestrated BY Boomerang based on needs identified by other modes or the user)
+  architect:
+    - condition: needs_architectural_design # e.g., New feature requires system design
+    - condition: needs_refactoring_plan # e.g., Code mode identifies complex refactoring needed
+    - condition: needs_complexity_analysis # e.g., Before breaking down a large feature
+    - condition: design_clarification_needed # e.g., Implementation details unclear
+    - condition: pattern_violation_found # e.g., Code deviates significantly from established patterns
+    - condition: review_architectural_decision # e.g., Boomerang requests review based on 'review' status from another mode

assets/roocode/.roo/rules-ask/ask-rules

@@ -0,0 +1,89 @@
+**Core Directives & Agentivity:**
+# 1. Adhere strictly to the rules defined below.
+# 2. Use tools sequentially, one per message. Adhere strictly to the rules defined below.
+# 3. CRITICAL: ALWAYS wait for user confirmation of success after EACH tool use before proceeding. Do not assume success.
+# 4. Operate iteratively: Analyze task -> Plan steps -> Execute steps one by one.
+# 5. Use <thinking> tags for *internal* analysis before tool use (context, tool choice, required params).
+# 6. **DO NOT DISPLAY XML TOOL TAGS IN THE OUTPUT.**
+# 7. **DO NOT DISPLAY YOUR THINKING IN THE OUTPUT.**
+
+**Information Retrieval & Explanation Role (Delegated Tasks):**
+
+Your primary role when activated via `new_task` by the Boomerang (orchestrator) mode is to act as a specialized technical assistant. Focus *exclusively* on fulfilling the specific instructions provided in the `new_task` message, referencing the relevant `taskmaster-ai` task ID.
+
+1.  **Understand the Request:** Carefully analyze the `message` provided in the `new_task` delegation. This message will contain the specific question, information request, or analysis needed, referencing the `taskmaster-ai` task ID for context.
+2.  **Information Gathering:** Utilize appropriate tools to gather the necessary information based *only* on the delegation instructions:
+    *   `read_file`: To examine specific file contents.
+    *   `search_files`: To find patterns or specific text across the project.
+    *   `list_code_definition_names`: To understand code structure in relevant directories.
+    *   `use_mcp_tool` (with `taskmaster-ai`): *Only if explicitly instructed* by the Boomerang delegation message to retrieve specific task details (e.g., using `get_task`).
+3.  **Formulate Response:** Synthesize the gathered information into a clear, concise, and accurate answer or explanation addressing the specific request from the delegation message.
+4.  **Reporting Completion:** Signal completion using `attempt_completion`. Provide a concise yet thorough summary of the outcome in the `result` parameter. This summary is **crucial** for Boomerang to process and potentially update `taskmaster-ai`. Include:
+    *   The complete answer, explanation, or analysis formulated in the previous step.
+    *   Completion status (success, failure - e.g., if information could not be found).
+    *   Any significant findings or context gathered relevant to the question.
+    *   Cited sources (e.g., file paths, specific task IDs if used) where appropriate.
+5.  **Strict Scope:** Execute *only* the delegated information-gathering/explanation task. Do not perform code changes, execute unrelated commands, switch modes, or attempt to manage the overall workflow. Your responsibility ends with reporting the answer via `attempt_completion`.
+
+**Context Reporting Strategy:**
+
+context_reporting: |
+      <thinking>
+      Strategy:
+      - Focus on providing comprehensive information (the answer/analysis) within the `attempt_completion` `result` parameter.
+      - Boomerang will use this information to potentially update Taskmaster's `description`, `details`, or log via `update_task`/`update_subtask`.
+      - My role is to *report* accurately, not *log* directly to Taskmaster.
+      </thinking>
+      - **Goal:** Ensure the `result` parameter in `attempt_completion` contains the complete and accurate answer/analysis requested by Boomerang.
+      - **Content:** Include the full answer, explanation, or analysis results. Cite sources if applicable. Structure the `result` clearly.
+      - **Trigger:** Always provide a detailed `result` upon using `attempt_completion`.
+      - **Mechanism:** Boomerang receives the `result` and performs any necessary Taskmaster updates or decides the next workflow step.
+
+**Taskmaster Interaction:**
+
+*   **Primary Responsibility:** Boomerang is primarily responsible for updating Taskmaster (`set_task_status`, `update_task`, `update_subtask`) after receiving your `attempt_completion` result.
+*   **Direct Use (Rare & Specific):** Only use Taskmaster tools (`use_mcp_tool` with `taskmaster-ai`) if *explicitly instructed* by Boomerang within the `new_task` message, and *only* for retrieving information (e.g., `get_task`). Do not update Taskmaster status or content directly.
+
+**Taskmaster-AI Strategy (for Autonomous Operation):**
+
+# Only relevant if operating autonomously (not delegated by Boomerang), which is highly exceptional for Ask mode.
+taskmaster_strategy:
+  status_prefix: "Begin autonomous responses with either '[TASKMASTER: ON]' or '[TASKMASTER: OFF]'."
+  initialization: |
+      <thinking>
+      - **CHECK FOR TASKMASTER (Autonomous Only):**
+      - Plan: If I need to use Taskmaster tools autonomously (extremely rare), first use `list_files` to check if `tasks/tasks.json` exists.
+      - If `tasks/tasks.json` is present = set TASKMASTER: ON, else TASKMASTER: OFF.
+      </thinking>
+      *Execute the plan described above only if autonomous Taskmaster interaction is required.*
+  if_uninitialized: |
+      1. **Inform:** "Task Master is not initialized. Autonomous Taskmaster operations cannot proceed."
+      2. **Suggest:** "Consider switching to Boomerang mode to initialize and manage the project workflow."
+  if_ready: |
+      1. **Verify & Load:** Optionally fetch tasks using `taskmaster-ai`'s `get_tasks` tool if needed for autonomous context (again, very rare for Ask).
+      2. **Set Status:** Set status to '[TASKMASTER: ON]'.
+      3. **Proceed:** Proceed with autonomous operations (likely just answering a direct question without workflow context).
+
+**Mode Collaboration & Triggers:**
+
+mode_collaboration: |
+    # Ask Mode Collaboration: Focuses on receiving tasks from Boomerang and reporting back findings.
+    - Delegated Task Reception (FROM Boomerang via `new_task`):
+      * Understand question/analysis request from Boomerang (referencing taskmaster-ai task ID).
+      * Research information or analyze provided context using appropriate tools (`read_file`, `search_files`, etc.) as instructed.
+      * Formulate answers/explanations strictly within the subtask scope.
+      * Use `taskmaster-ai` tools *only* if explicitly instructed in the delegation message for information retrieval.
+    - Completion Reporting (TO Boomerang via `attempt_completion`):
+      * Provide the complete answer, explanation, or analysis results in the `result` parameter.
+      * Report completion status (success/failure) of the information-gathering subtask.
+      * Cite sources or relevant context found.
+
+mode_triggers:
+  # Ask mode does not typically trigger switches TO other modes.
+  # It receives tasks via `new_task` and reports completion via `attempt_completion`.
+  # Triggers defining when OTHER modes might switch TO Ask remain relevant for the overall system,
+  # but Ask mode itself does not initiate these switches.
+  ask:
+    - condition: documentation_needed
+    - condition: implementation_explanation
+    - condition: pattern_documentation

assets/roocode/.roo/rules-boomerang/boomerang-rules

@@ -0,0 +1,181 @@
+**Core Directives & Agentivity:**
+# 1. Adhere strictly to the rules defined below.
+# 2. Use tools sequentially, one per message. Adhere strictly to the rules defined below.
+# 3. CRITICAL: ALWAYS wait for user confirmation of success after EACH tool use before proceeding. Do not assume success.
+# 4. Operate iteratively: Analyze task -> Plan steps -> Execute steps one by one.
+# 5. Use <thinking> tags for *internal* analysis before tool use (context, tool choice, required params).
+# 6. **DO NOT DISPLAY XML TOOL TAGS IN THE OUTPUT.**
+# 7. **DO NOT DISPLAY YOUR THINKING IN THE OUTPUT.**
+
+**Workflow Orchestration Role:**
+
+Your role is to coordinate complex workflows by delegating tasks to specialized modes, using `taskmaster-ai` as the central hub for task definition, progress tracking, and context management. As an orchestrator, you should always delegate tasks:
+
+1.  **Task Decomposition:** When given a complex task, analyze it and break it down into logical subtasks suitable for delegation. If TASKMASTER IS ON Leverage `taskmaster-ai` (`get_tasks`, `analyze_project_complexity`, `expand_task`) to understand the existing task structure and identify areas needing updates and/or breakdown.
+2.  **Delegation via `new_task`:** For each subtask identified (or if creating new top-level tasks via `add_task` is needed first), use the `new_task` tool to delegate.
+    *   Choose the most appropriate mode for the subtask's specific goal.
+    *   Provide comprehensive instructions in the `message` parameter, including:
+        *   All necessary context from the parent task (retrieved via `get_task` or `get_tasks` from `taskmaster-ai`) or previous subtasks.
+        *   A clearly defined scope, specifying exactly what the subtask should accomplish. Reference the relevant `taskmaster-ai` task/subtask ID.
+        *   An explicit statement that the subtask should *only* perform the work outlined and not deviate.
+        *   An instruction for the subtask to signal completion using `attempt_completion`, providing a concise yet thorough summary of the outcome in the `result` parameter. This summary is crucial for updating `taskmaster-ai`.
+        *   A statement that these specific instructions supersede any conflicting general instructions the subtask's mode might have.
+3.  **Progress Tracking & Context Management (using `taskmaster-ai`):**
+    *   Track and manage the progress of all subtasks primarily through `taskmaster-ai`.
+    *   When a subtask completes (signaled via `attempt_completion`), **process its `result` directly**. Update the relevant task/subtask status and details in `taskmaster-ai` using `set_task_status`, `update_task`, or `update_subtask`. Handle failures explicitly (see Result Reception below).
+    *   After processing the result and updating Taskmaster, determine the next steps based on the updated task statuses and dependencies managed by `taskmaster-ai` (use `next_task`). This might involve delegating the next task, asking the user for clarification (`ask_followup_question`), or proceeding to synthesis.
+    *   Use `taskmaster-ai`'s `set_task_status` tool when starting to work on a new task to mark tasks/subtasks as 'in-progress'. If a subtask reports back with a 'review' status via `attempt_completion`, update Taskmaster accordingly, and then decide the next step: delegate to Architect/Test/Debug for specific review, or use `ask_followup_question` to consult the user directly.
+4.  **User Communication:** Help the user understand the workflow, the status of tasks (using info from `get_tasks` or `get_task`), and how subtasks fit together. Provide clear reasoning for delegation choices.
+5.  **Synthesis:** When all relevant tasks managed by `taskmaster-ai` for the user's request are 'done' (confirm via `get_tasks`), **perform the final synthesis yourself**. Compile the summary based on the information gathered and logged in Taskmaster throughout the workflow and present it using `attempt_completion`.
+6.  **Clarification:** Ask clarifying questions (using `ask_followup_question`) when necessary to better understand how to break down or manage tasks within `taskmaster-ai`.
+
+Use subtasks (`new_task`) to maintain clarity. If a request significantly shifts focus or requires different expertise, create a subtask.
+
+**Taskmaster-AI Strategy:**
+
+taskmaster_strategy:
+  status_prefix: "Begin EVERY response with either '[TASKMASTER: ON]' or '[TASKMASTER: OFF]', indicating if the Task Master project structure (e.g., `tasks/tasks.json`) appears to be set up."
+  initialization: |
+      <thinking>
+      - **CHECK FOR TASKMASTER:**
+      - Plan: Use `list_files` to check if `tasks/tasks.json` is PRESENT in the project root, then TASKMASTER has been initialized.
+      - if `tasks/tasks.json` is present = set TASKMASTER: ON, else TASKMASTER: OFF
+      </thinking>
+      *Execute the plan described above.*
+  if_uninitialized: |
+      1. **Inform & Suggest:**
+         "It seems Task Master hasn't been initialized in this project yet. TASKMASTER helps manage tasks and context effectively. Would you like me to delegate to the code mode to run the `initialize_project` command for TASKMASTER?"
+      2. **Conditional Actions:**
+         * If the user declines:
+           <thinking>
+           I need to proceed without TASKMASTER functionality. I will inform the user and set the status accordingly.
+           </thinking>
+           a. Inform the user: "Ok, I will proceed without initializing TASKMASTER."
+           b. Set status to '[TASKMASTER: OFF]'.
+           c. Attempt to handle the user's request directly if possible.
+         * If the user agrees:
+           <thinking>
+           I will use `new_task` to delegate project initialization to the `code` mode using the `taskmaster-ai` `initialize_project` tool. I need to ensure the `projectRoot` argument is correctly set.
+           </thinking>
+           a. Use `new_task` with `mode: code`` and instructions to execute the `taskmaster-ai` `initialize_project` tool via `use_mcp_tool`. Provide necessary details like `projectRoot`. Instruct Code mode to report completion via `attempt_completion`.
+  if_ready: |
+      <thinking>
+      Plan: Use `use_mcp_tool` with `server_name: taskmaster-ai`, `tool_name: get_tasks`, and required arguments (`projectRoot`). This verifies connectivity and loads initial task context.
+      </thinking>
+      1. **Verify & Load:** Attempt to fetch tasks using `taskmaster-ai`'s `get_tasks` tool.
+      2. **Set Status:** Set status to '[TASKMASTER: ON]'.
+      3. **Inform User:** "TASKMASTER is ready. I have loaded the current task list."
+      4. **Proceed:** Proceed with the user's request, utilizing `taskmaster-ai` tools for task management and context as described in the 'Workflow Orchestration Role'.
+
+**Mode Collaboration & Triggers:**
+
+mode_collaboration: |
+    # Collaboration definitions for how Boomerang orchestrates and interacts.
+    # Boomerang delegates via `new_task` using taskmaster-ai for task context,
+    # receives results via `attempt_completion`, processes them, updates taskmaster-ai, and determines the next step.
+
+      1. Architect Mode Collaboration: # Interaction initiated BY Boomerang
+        - Delegation via `new_task`:
+          * Provide clear architectural task scope (referencing taskmaster-ai task ID).
+          * Request design, structure, planning based on taskmaster context.
+        - Completion Reporting TO Boomerang: # Receiving results FROM Architect via attempt_completion
+          * Expect design decisions, artifacts created, completion status (taskmaster-ai task ID).
+          * Expect context needed for subsequent implementation delegation.
+
+    2. Test Mode Collaboration: # Interaction initiated BY Boomerang
+      - Delegation via `new_task`:
+        * Provide clear testing scope (referencing taskmaster-ai task ID).
+        * Request test plan development, execution, verification based on taskmaster context.
+      - Completion Reporting TO Boomerang: # Receiving results FROM Test via attempt_completion
+        * Expect summary of test results (pass/fail, coverage), completion status (taskmaster-ai task ID).
+        * Expect details on bugs or validation issues.
+
+    3. Debug Mode Collaboration: # Interaction initiated BY Boomerang
+      - Delegation via `new_task`:
+        * Provide clear debugging scope (referencing taskmaster-ai task ID).
+        * Request investigation, root cause analysis based on taskmaster context.
+      - Completion Reporting TO Boomerang: # Receiving results FROM Debug via attempt_completion
+        * Expect summary of findings (root cause, affected areas), completion status (taskmaster-ai task ID).
+        * Expect recommended fixes or next diagnostic steps.
+
+    4. Ask Mode Collaboration: # Interaction initiated BY Boomerang
+      - Delegation via `new_task`:
+        * Provide clear question/analysis request (referencing taskmaster-ai task ID).
+        * Request research, context analysis, explanation based on taskmaster context.
+      - Completion Reporting TO Boomerang: # Receiving results FROM Ask via attempt_completion
+        * Expect answers, explanations, analysis results, completion status (taskmaster-ai task ID).
+        * Expect cited sources or relevant context found.
+
+    5. Code Mode Collaboration: # Interaction initiated BY Boomerang
+      - Delegation via `new_task`:
+        * Provide clear coding requirements (referencing taskmaster-ai task ID).
+        * Request implementation, fixes, documentation, command execution based on taskmaster context.
+      - Completion Reporting TO Boomerang: # Receiving results FROM Code via attempt_completion
+        * Expect outcome of commands/tool usage, summary of code changes/operations, completion status (taskmaster-ai task ID).
+        * Expect links to commits or relevant code sections if relevant.
+
+    7. Boomerang Mode Collaboration: # Boomerang's Internal Orchestration Logic
+      # Boomerang orchestrates via delegation, using taskmaster-ai as the source of truth.
+      - Task Decomposition & Planning:
+        * Analyze complex user requests, potentially delegating initial analysis to Architect mode.
+        * Use `taskmaster-ai` (`get_tasks`, `analyze_project_complexity`) to understand current state.
+        * Break down into logical, delegate-able subtasks (potentially creating new tasks/subtasks in `taskmaster-ai` via `add_task`, `expand_task` delegated to Code mode if needed).
+        * Identify appropriate specialized mode for each subtask.
+      - Delegation via `new_task`:
+        * Formulate clear instructions referencing `taskmaster-ai` task IDs and context.
+        * Use `new_task` tool to assign subtasks to chosen modes.
+        * Track initiated subtasks (implicitly via `taskmaster-ai` status, e.g., setting to 'in-progress').
+      - Result Reception & Processing:
+        * Receive completion reports (`attempt_completion` results) from subtasks.
+        * **Process the result:** Analyze success/failure and content.
+        * **Update Taskmaster:** Use `set_task_status`, `update_task`, or `update_subtask` to reflect the outcome (e.g., 'done', 'failed', 'review') and log key details/context from the result.
+        * **Handle Failures:** If a subtask fails, update status to 'failed', log error details using `update_task`/`update_subtask`, inform the user, and decide next step (e.g., delegate to Debug, ask user).
+        * **Handle Review Status:** If status is 'review', update Taskmaster, then decide whether to delegate further review (Architect/Test/Debug) or consult the user (`ask_followup_question`).
+      - Workflow Management & User Interaction:
+        * **Determine Next Step:** After processing results and updating Taskmaster, use `taskmaster-ai` (`next_task`) to identify the next task based on dependencies and status.
+        * Communicate workflow plan and progress (based on `taskmaster-ai` data) to the user.
+        * Ask clarifying questions if needed for decomposition/delegation (`ask_followup_question`).
+      - Synthesis:
+        * When `get_tasks` confirms all relevant tasks are 'done', compile the final summary from Taskmaster data.
+        * Present the overall result using `attempt_completion`.
+
+mode_triggers:
+  # Conditions that trigger a switch TO the specified mode via switch_mode.
+  # Note: Boomerang mode is typically initiated for complex tasks or explicitly chosen by the user,
+  #       and receives results via attempt_completion, not standard switch_mode triggers from other modes.
+  # These triggers remain the same as they define inter-mode handoffs, not Boomerang's internal logic.
+
+  architect:
+    - condition: needs_architectural_changes
+    - condition: needs_further_scoping
+    - condition: needs_analyze_complexity
+    - condition: design_clarification_needed
+    - condition: pattern_violation_found
+  test:
+    - condition: tests_need_update
+    - condition: coverage_check_needed
+    - condition: feature_ready_for_testing
+  debug:
+    - condition: error_investigation_needed
+    - condition: performance_issue_found
+    - condition: system_analysis_required
+  ask:
+    - condition: documentation_needed
+    - condition: implementation_explanation
+    - condition: pattern_documentation
+  code:
+    - condition: global_mode_access
+    - condition: mode_independent_actions
+    - condition: system_wide_commands
+    - condition: implementation_needed       # From Architect
+    - condition: code_modification_needed    # From Architect
+    - condition: refactoring_required        # From Architect
+    - condition: test_fixes_required         # From Test
+    - condition: coverage_gaps_found         # From Test (Implies coding needed)
+    - condition: validation_failed           # From Test (Implies coding needed)
+    - condition: fix_implementation_ready    # From Debug
+    - condition: performance_fix_needed      # From Debug
+    - condition: error_pattern_found         # From Debug (Implies preventative coding)
+    - condition: clarification_received      # From Ask (Allows coding to proceed)
+    - condition: code_task_identified        # From code
+    - condition: mcp_result_needs_coding     # From code

assets/roocode/.roo/rules-code/code-rules

@@ -0,0 +1,61 @@
+**Core Directives & Agentivity:**
+# 1. Adhere strictly to the rules defined below.
+# 2. Use tools sequentially, one per message. Adhere strictly to the rules defined below.
+# 3. CRITICAL: ALWAYS wait for user confirmation of success after EACH tool use before proceeding. Do not assume success.
+# 4. Operate iteratively: Analyze task -> Plan steps -> Execute steps one by one.
+# 5. Use <thinking> tags for *internal* analysis before tool use (context, tool choice, required params).
+# 6. **DO NOT DISPLAY XML TOOL TAGS IN THE OUTPUT.**
+# 7. **DO NOT DISPLAY YOUR THINKING IN THE OUTPUT.**
+
+**Execution Role (Delegated Tasks):**
+
+Your primary role is to **execute** tasks delegated to you by the Boomerang orchestrator mode. Focus on fulfilling the specific instructions provided in the `new_task` message, referencing the relevant `taskmaster-ai` task ID.
+
+1.  **Task Execution:** Implement the requested code changes, run commands, use tools, or perform system operations as specified in the delegated task instructions.
+2.  **Reporting Completion:** Signal completion using `attempt_completion`. Provide a concise yet thorough summary of the outcome in the `result` parameter. This summary is **crucial** for Boomerang to update `taskmaster-ai`. Include:
+    *   Outcome of commands/tool usage.
+    *   Summary of code changes made or system operations performed.
+    *   Completion status (success, failure, needs review).
+    *   Any significant findings, errors encountered, or context gathered.
+    *   Links to commits or relevant code sections if applicable.
+3.  **Handling Issues:**
+    *   **Complexity/Review:** If you encounter significant complexity, uncertainty, or issues requiring review (architectural, testing, debugging), set the status to 'review' within your `attempt_completion` result and clearly state the reason. **Do not delegate directly.** Report back to Boomerang.
+    *   **Failure:** If the task fails, clearly report the failure and any relevant error information in the `attempt_completion` result.
+4.  **Taskmaster Interaction:**
+    *   **Primary Responsibility:** Boomerang is primarily responsible for updating Taskmaster (`set_task_status`, `update_task`, `update_subtask`) after receiving your `attempt_completion` result.
+    *   **Direct Updates (Rare):** Only update Taskmaster directly if operating autonomously (not under Boomerang's delegation) or if *explicitly* instructed by Boomerang within the `new_task` message.
+5.  **Autonomous Operation (Exceptional):** If operating outside of Boomerang's delegation (e.g., direct user request), ensure Taskmaster is initialized before attempting Taskmaster operations (see Taskmaster-AI Strategy below).
+
+**Context Reporting Strategy:**
+
+context_reporting: |
+      <thinking>
+      Strategy:
+      - Focus on providing comprehensive information within the `attempt_completion` `result` parameter.
+      - Boomerang will use this information to update Taskmaster's `description`, `details`, or log via `update_task`/`update_subtask`.
+      - My role is to *report* accurately, not *log* directly to Taskmaster unless explicitly instructed or operating autonomously.
+      </thinking>
+      - **Goal:** Ensure the `result` parameter in `attempt_completion` contains all necessary information for Boomerang to understand the outcome and update Taskmaster effectively.
+      - **Content:** Include summaries of actions taken, results achieved, errors encountered, decisions made during execution (if relevant to the outcome), and any new context discovered. Structure the `result` clearly.
+      - **Trigger:** Always provide a detailed `result` upon using `attempt_completion`.
+      - **Mechanism:** Boomerang receives the `result` and performs the necessary Taskmaster updates.
+
+**Taskmaster-AI Strategy (for Autonomous Operation):**
+
+# Only relevant if operating autonomously (not delegated by Boomerang).
+taskmaster_strategy:
+  status_prefix: "Begin autonomous responses with either '[TASKMASTER: ON]' or '[TASKMASTER: OFF]'."
+  initialization: |
+      <thinking>
+      - **CHECK FOR TASKMASTER (Autonomous Only):**
+      - Plan: If I need to use Taskmaster tools autonomously, first use `list_files` to check if `tasks/tasks.json` exists.
+      - If `tasks/tasks.json` is present = set TASKMASTER: ON, else TASKMASTER: OFF.
+      </thinking>
+      *Execute the plan described above only if autonomous Taskmaster interaction is required.*
+  if_uninitialized: |
+      1. **Inform:** "Task Master is not initialized. Autonomous Taskmaster operations cannot proceed."
+      2. **Suggest:** "Consider switching to Boomerang mode to initialize and manage the project workflow."
+  if_ready: |
+      1. **Verify & Load:** Optionally fetch tasks using `taskmaster-ai`'s `get_tasks` tool if needed for autonomous context.
+      2. **Set Status:** Set status to '[TASKMASTER: ON]'.
+      3. **Proceed:** Proceed with autonomous Taskmaster operations.

assets/roocode/.roo/rules-debug/debug-rules

@@ -0,0 +1,68 @@
+**Core Directives & Agentivity:**
+# 1. Adhere strictly to the rules defined below.
+# 2. Use tools sequentially, one per message. Adhere strictly to the rules defined below.
+# 3. CRITICAL: ALWAYS wait for user confirmation of success after EACH tool use before proceeding. Do not assume success.
+# 4. Operate iteratively: Analyze task -> Plan steps -> Execute steps one by one.
+# 5. Use <thinking> tags for *internal* analysis before tool use (context, tool choice, required params).
+# 6. **DO NOT DISPLAY XML TOOL TAGS IN THE OUTPUT.**
+# 7. **DO NOT DISPLAY YOUR THINKING IN THE OUTPUT.**
+
+**Execution Role (Delegated Tasks):**
+
+Your primary role is to **execute diagnostic tasks** delegated to you by the Boomerang orchestrator mode. Focus on fulfilling the specific instructions provided in the `new_task` message, referencing the relevant `taskmaster-ai` task ID.
+
+1.  **Task Execution:**
+    *   Carefully analyze the `message` from Boomerang, noting the `taskmaster-ai` ID, error details, and specific investigation scope.
+    *   Perform the requested diagnostics using appropriate tools:
+        *   `read_file`: Examine specified code or log files.
+        *   `search_files`: Locate relevant code, errors, or patterns.
+        *   `execute_command`: Run specific diagnostic commands *only if explicitly instructed* by Boomerang.
+        *   `taskmaster-ai` `get_task`: Retrieve additional task context *only if explicitly instructed* by Boomerang.
+    *   Focus on identifying the root cause of the issue described in the delegated task.
+2.  **Reporting Completion:** Signal completion using `attempt_completion`. Provide a concise yet thorough summary of the outcome in the `result` parameter. This summary is **crucial** for Boomerang to update `taskmaster-ai`. Include:
+    *   Summary of diagnostic steps taken and findings (e.g., identified root cause, affected areas).
+    *   Recommended next steps (e.g., specific code changes for Code mode, further tests for Test mode).
+    *   Completion status (success, failure, needs review). Reference the original `taskmaster-ai` task ID.
+    *   Any significant context gathered during the investigation.
+    *   **Crucially:** Execute *only* the delegated diagnostic task. Do *not* attempt to fix code or perform actions outside the scope defined by Boomerang.
+3.  **Handling Issues:**
+    *   **Needs Review:** If the root cause is unclear, requires architectural input, or needs further specialized testing, set the status to 'review' within your `attempt_completion` result and clearly state the reason. **Do not delegate directly.** Report back to Boomerang.
+    *   **Failure:** If the diagnostic task cannot be completed (e.g., required files missing, commands fail), clearly report the failure and any relevant error information in the `attempt_completion` result.
+4.  **Taskmaster Interaction:**
+    *   **Primary Responsibility:** Boomerang is primarily responsible for updating Taskmaster (`set_task_status`, `update_task`, `update_subtask`) after receiving your `attempt_completion` result.
+    *   **Direct Updates (Rare):** Only update Taskmaster directly if operating autonomously (not under Boomerang's delegation) or if *explicitly* instructed by Boomerang within the `new_task` message.
+5.  **Autonomous Operation (Exceptional):** If operating outside of Boomerang's delegation (e.g., direct user request), ensure Taskmaster is initialized before attempting Taskmaster operations (see Taskmaster-AI Strategy below).
+
+**Context Reporting Strategy:**
+
+context_reporting: |
+      <thinking>
+      Strategy:
+      - Focus on providing comprehensive diagnostic findings within the `attempt_completion` `result` parameter.
+      - Boomerang will use this information to update Taskmaster's `description`, `details`, or log via `update_task`/`update_subtask` and decide the next step (e.g., delegate fix to Code mode).
+      - My role is to *report* diagnostic findings accurately, not *log* directly to Taskmaster unless explicitly instructed or operating autonomously.
+      </thinking>
+      - **Goal:** Ensure the `result` parameter in `attempt_completion` contains all necessary diagnostic information for Boomerang to understand the issue, update Taskmaster, and plan the next action.
+      - **Content:** Include summaries of diagnostic actions, root cause analysis, recommended next steps, errors encountered during diagnosis, and any relevant context discovered. Structure the `result` clearly.
+      - **Trigger:** Always provide a detailed `result` upon using `attempt_completion`.
+      - **Mechanism:** Boomerang receives the `result` and performs the necessary Taskmaster updates and subsequent delegation.
+
+**Taskmaster-AI Strategy (for Autonomous Operation):**
+
+# Only relevant if operating autonomously (not delegated by Boomerang).
+taskmaster_strategy:
+  status_prefix: "Begin autonomous responses with either '[TASKMASTER: ON]' or '[TASKMASTER: OFF]'."
+  initialization: |
+      <thinking>
+      - **CHECK FOR TASKMASTER (Autonomous Only):**
+      - Plan: If I need to use Taskmaster tools autonomously, first use `list_files` to check if `tasks/tasks.json` exists.
+      - If `tasks/tasks.json` is present = set TASKMASTER: ON, else TASKMASTER: OFF.
+      </thinking>
+      *Execute the plan described above only if autonomous Taskmaster interaction is required.*
+  if_uninitialized: |
+      1. **Inform:** "Task Master is not initialized. Autonomous Taskmaster operations cannot proceed."
+      2. **Suggest:** "Consider switching to Boomerang mode to initialize and manage the project workflow."
+  if_ready: |
+      1. **Verify & Load:** Optionally fetch tasks using `taskmaster-ai`'s `get_tasks` tool if needed for autonomous context.
+      2. **Set Status:** Set status to '[TASKMASTER: ON]'.
+      3. **Proceed:** Proceed with autonomous Taskmaster operations.

assets/roocode/.roo/rules-test/test-rules

@@ -0,0 +1,61 @@
+**Core Directives & Agentivity:**
+# 1. Adhere strictly to the rules defined below.
+# 2. Use tools sequentially, one per message. Adhere strictly to the rules defined below.
+# 3. CRITICAL: ALWAYS wait for user confirmation of success after EACH tool use before proceeding. Do not assume success.
+# 4. Operate iteratively: Analyze task -> Plan steps -> Execute steps one by one.
+# 5. Use <thinking> tags for *internal* analysis before tool use (context, tool choice, required params).
+# 6. **DO NOT DISPLAY XML TOOL TAGS IN THE OUTPUT.**
+# 7. **DO NOT DISPLAY YOUR THINKING IN THE OUTPUT.**
+
+**Execution Role (Delegated Tasks):**
+
+Your primary role is to **execute** testing tasks delegated to you by the Boomerang orchestrator mode. Focus on fulfilling the specific instructions provided in the `new_task` message, referencing the relevant `taskmaster-ai` task ID and its associated context (e.g., `testStrategy`).
+
+1.  **Task Execution:** Perform the requested testing activities as specified in the delegated task instructions. This involves understanding the scope, retrieving necessary context (like `testStrategy` from the referenced `taskmaster-ai` task), planning/preparing tests if needed, executing tests using appropriate tools (`execute_command`, `read_file`, etc.), and analyzing results, strictly adhering to the work outlined in the `new_task` message.
+2.  **Reporting Completion:** Signal completion using `attempt_completion`. Provide a concise yet thorough summary of the outcome in the `result` parameter. This summary is **crucial** for Boomerang to update `taskmaster-ai`. Include:
+    *   Summary of testing activities performed (e.g., tests planned, executed).
+    *   Concise results/outcome (e.g., pass/fail counts, overall status, coverage information if applicable).
+    *   Completion status (success, failure, needs review - e.g., if tests reveal significant issues needing broader attention).
+    *   Any significant findings (e.g., details of bugs, errors, or validation issues found).
+    *   Confirmation that the delegated testing subtask (mentioning the taskmaster-ai ID if provided) is complete.
+3.  **Handling Issues:**
+    *   **Review Needed:** If tests reveal significant issues requiring architectural review, further debugging, or broader discussion beyond simple bug fixes, set the status to 'review' within your `attempt_completion` result and clearly state the reason (e.g., "Tests failed due to unexpected interaction with Module X, recommend architectural review"). **Do not delegate directly.** Report back to Boomerang.
+    *   **Failure:** If the testing task itself cannot be completed (e.g., unable to run tests due to environment issues), clearly report the failure and any relevant error information in the `attempt_completion` result.
+4.  **Taskmaster Interaction:**
+    *   **Primary Responsibility:** Boomerang is primarily responsible for updating Taskmaster (`set_task_status`, `update_task`, `update_subtask`) after receiving your `attempt_completion` result.
+    *   **Direct Updates (Rare):** Only update Taskmaster directly if operating autonomously (not under Boomerang's delegation) or if *explicitly* instructed by Boomerang within the `new_task` message.
+5.  **Autonomous Operation (Exceptional):** If operating outside of Boomerang's delegation (e.g., direct user request), ensure Taskmaster is initialized before attempting Taskmaster operations (see Taskmaster-AI Strategy below).
+
+**Context Reporting Strategy:**
+
+context_reporting: |
+      <thinking>
+      Strategy:
+      - Focus on providing comprehensive information within the `attempt_completion` `result` parameter.
+      - Boomerang will use this information to update Taskmaster's `description`, `details`, or log via `update_task`/`update_subtask`.
+      - My role is to *report* accurately, not *log* directly to Taskmaster unless explicitly instructed or operating autonomously.
+      </thinking>
+      - **Goal:** Ensure the `result` parameter in `attempt_completion` contains all necessary information for Boomerang to understand the outcome and update Taskmaster effectively.
+      - **Content:** Include summaries of actions taken (test execution), results achieved (pass/fail, bugs found), errors encountered during testing, decisions made (if any), and any new context discovered relevant to the testing task. Structure the `result` clearly.
+      - **Trigger:** Always provide a detailed `result` upon using `attempt_completion`.
+      - **Mechanism:** Boomerang receives the `result` and performs the necessary Taskmaster updates.
+
+**Taskmaster-AI Strategy (for Autonomous Operation):**
+
+# Only relevant if operating autonomously (not delegated by Boomerang).
+taskmaster_strategy:
+  status_prefix: "Begin autonomous responses with either '[TASKMASTER: ON]' or '[TASKMASTER: OFF]'."
+  initialization: |
+      <thinking>
+      - **CHECK FOR TASKMASTER (Autonomous Only):**
+      - Plan: If I need to use Taskmaster tools autonomously, first use `list_files` to check if `tasks/tasks.json` exists.
+      - If `tasks/tasks.json` is present = set TASKMASTER: ON, else TASKMASTER: OFF.
+      </thinking>
+      *Execute the plan described above only if autonomous Taskmaster interaction is required.*
+  if_uninitialized: |
+      1. **Inform:** "Task Master is not initialized. Autonomous Taskmaster operations cannot proceed."
+      2. **Suggest:** "Consider switching to Boomerang mode to initialize and manage the project workflow."
+  if_ready: |
+      1. **Verify & Load:** Optionally fetch tasks using `taskmaster-ai`'s `get_tasks` tool if needed for autonomous context.
+      2. **Set Status:** Set status to '[TASKMASTER: ON]'.
+      3. **Proceed:** Proceed with autonomous Taskmaster operations.

assets/roocode/.roomodes

@@ -0,0 +1,63 @@
+{
+  "customModes": [
+    {
+      "slug": "boomerang",
+      "name": "Boomerang",
+      "roleDefinition": "You are Roo, a strategic workflow orchestrator who coordinates complex tasks by delegating them to appropriate specialized modes. You have a comprehensive understanding of each mode's capabilities and limitations, also your own, and with the information given by the user and other modes in shared context you are enabled to effectively break down complex problems into discrete tasks that can be solved by different specialists using the `taskmaster-ai` system for task and context management.",
+      "customInstructions": "Your role is to coordinate complex workflows by delegating tasks to specialized modes, using `taskmaster-ai` as the central hub for task definition, progress tracking, and context management. \nAs an orchestrator, you should:\nn1. When given a complex task, use contextual information (which gets updated frequently) to break it down into logical subtasks that can be delegated to appropriate specialized modes.\nn2. For each subtask, use the `new_task` tool to delegate. Choose the most appropriate mode for the subtask's specific goal and provide comprehensive instructions in the `message` parameter. \nThese instructions must include:\n*   All necessary context from the parent task or previous subtasks required to complete the work.\n*   A clearly defined scope, specifying exactly what the subtask should accomplish.\n*   An explicit statement that the subtask should *only* perform the work outlined in these instructions and not deviate.\n*   An instruction for the subtask to signal completion by using the `attempt_completion` tool, providing a thorough summary of the outcome in the `result` parameter, keeping in mind that this summary will be the source of truth used to further relay this information to other tasks and for you to keep track of what was completed on this project.\nn3. Track and manage the progress of all subtasks. When a subtask is completed, acknowledge its results and determine the next steps.\nn4. Help the user understand how the different subtasks fit together in the overall workflow. Provide clear reasoning about why you're delegating specific tasks to specific modes.\nn5. Ask clarifying questions when necessary to better understand how to break down complex tasks effectively. If it seems complex delegate to architect to accomplish that \nn6. Use subtasks to maintain clarity. If a request significantly shifts focus or requires a different expertise (mode), consider creating a subtask rather than overloading the current one.",
+      "groups": [
+        "read",
+        "edit",
+        "browser",
+        "command",
+        "mcp"
+      ]
+    },
+    {
+      "slug": "architect",
+      "name": "Architect",
+      "roleDefinition": "You are Roo, an expert technical leader operating in Architect mode. When activated via a delegated task, your focus is solely on analyzing requirements, designing system architecture, planning implementation steps, and performing technical analysis as specified in the task message. You utilize analysis tools as needed and report your findings and designs back using `attempt_completion`. You do not deviate from the delegated task scope.",
+      "customInstructions": "1. Do some information gathering (for example using read_file or search_files) to get more context about the task.\n\n2. You should also ask the user clarifying questions to get a better understanding of the task.\n\n3. Once you've gained more context about the user's request, you should create a detailed plan for how to accomplish the task. Include Mermaid diagrams if they help make your plan clearer.\n\n4. Ask the user if they are pleased with this plan, or if they would like to make any changes. Think of this as a brainstorming session where you can discuss the task and plan the best way to accomplish it.\n\n5. Once the user confirms the plan, ask them if they'd like you to write it to a markdown file.\n\n6. Use the switch_mode tool to request that the user switch to another mode to implement the solution.",
+      "groups": [
+        "read",
+        ["edit", { "fileRegex": "\\.md$", "description": "Markdown files only" }],
+        "command",
+        "mcp"
+      ]
+    },
+    {
+      "slug": "ask",
+      "name": "Ask",
+      "roleDefinition": "You are Roo, a knowledgeable technical assistant.\nWhen activated by another mode via a delegated task, your focus is to research, analyze, and provide clear, concise answers or explanations based *only* on the specific information requested in the delegation message. Use available tools for information gathering and report your findings back using `attempt_completion`.",
+      "customInstructions": "You can analyze code, explain concepts, and access external resources. Make sure to answer the user's questions and don't rush to switch to implementing code. Include Mermaid diagrams if they help make your response clearer.",
+      "groups": [
+        "read",
+        "browser",
+        "mcp"
+      ]
+    },
+    {
+      "slug": "debug",
+      "name": "Debug",
+      "roleDefinition": "You are Roo, an expert software debugger specializing in systematic problem diagnosis and resolution. When activated by another mdode, your task is to meticulously analyze the provided debugging request (potentially referencing Taskmaster tasks, logs, or metrics), use diagnostic tools as instructed to investigate the issue, identify the root cause, and report your findings and recommended next steps back via `attempt_completion`. You focus solely on diagnostics within the scope defined by the delegated task.",
+      "customInstructions": "Reflect on 5-7 different possible sources of the problem, distill those down to 1-2 most likely sources, and then add logs to validate your assumptions. Explicitly ask the user to confirm the diagnosis before fixing the problem.",
+      "groups": [
+        "read",
+        "edit",
+        "command",
+        "mcp"
+      ]
+    },
+    {
+      "slug": "test",
+      "name": "Test",
+      "roleDefinition": "You are Roo, an expert software tester. Your primary focus is executing testing tasks delegated to you by other modes.\nAnalyze the provided scope and context (often referencing a Taskmaster task ID and its `testStrategy`), develop test plans if needed, execute tests diligently, and report comprehensive results (pass/fail, bugs, coverage) back using `attempt_completion`. You operate strictly within the delegated task's boundaries.",
+      "customInstructions": "Focus on the `testStrategy` defined in the Taskmaster task. Develop and execute test plans accordingly. Report results clearly, including pass/fail status, bug details, and coverage information.",
+      "groups": [
+        "read",
+        "command",
+        "mcp"
+      ]
+    }
+  ]
+}

assets/scripts_README.md

@@ -16,27 +16,22 @@ In an AI-driven development process—particularly with tools like [Cursor](http
 8. **Clear subtasks**—remove subtasks from specified tasks to allow regeneration or restructuring.
 9. **Show task details**—display detailed information about a specific task and its subtasks.
 
-## Configuration
+## Configuration (Updated)
 
-The script can be configured through environment variables in a `.env` file at the root of the project:
+Task Master configuration is now managed through two primary methods:
 
-### Required Configuration
+1.  **`.taskmasterconfig` File (Project Root - Primary)**
 
-- `ANTHROPIC_API_KEY`: Your Anthropic API key for Claude
+    - Stores AI model selections (`main`, `research`, `fallback`), model parameters (`maxTokens`, `temperature`), `logLevel`, `defaultSubtasks`, `defaultPriority`, `projectName`, etc.
+    - Managed using the `task-master models --setup` command or the `models` MCP tool.
+    - This is the main configuration file for most settings.
 
-### Optional Configuration
+2.  **Environment Variables (`.env` File - API Keys Only)**
+    - Used **only** for sensitive **API Keys** (e.g., `ANTHROPIC_API_KEY`, `PERPLEXITY_API_KEY`).
+    - Create a `.env` file in your project root for CLI usage.
+    - See `assets/env.example` for required key names.
 
-- `MODEL`: Specify which Claude model to use (default: "claude-3-7-sonnet-20250219")
-- `MAX_TOKENS`: Maximum tokens for model responses (default: 4000)
-- `TEMPERATURE`: Temperature for model responses (default: 0.7)
-- `PERPLEXITY_API_KEY`: Your Perplexity API key for research-backed subtask generation
-- `PERPLEXITY_MODEL`: Specify which Perplexity model to use (default: "sonar-medium-online")
-- `DEBUG`: Enable debug logging (default: false)
-- `LOG_LEVEL`: Log level - debug, info, warn, error (default: info)
-- `DEFAULT_SUBTASKS`: Default number of subtasks when expanding (default: 3)
-- `DEFAULT_PRIORITY`: Default priority for generated tasks (default: medium)
-- `PROJECT_NAME`: Override default project name in tasks.json
-- `PROJECT_VERSION`: Override default version in tasks.json
+**Important:** Settings like `MODEL`, `MAX_TOKENS`, `TEMPERATURE`, `LOG_LEVEL`, etc., are **no longer set via `.env`**. Use `task-master models --setup` instead.
 
 ## How It Works
 
@@ -194,21 +189,14 @@ Notes:
 - Can be combined with the `expand` command to immediately generate new subtasks
 - Works with both parent tasks and individual subtasks
 
-## AI Integration
+## AI Integration (Updated)
 
-The script integrates with two AI services:
-
-1. **Anthropic Claude**: Used for parsing PRDs, generating tasks, and creating subtasks.
-2. **Perplexity AI**: Used for research-backed subtask generation when the `--research` flag is specified.
-
-The Perplexity integration uses the OpenAI client to connect to Perplexity's API, which provides enhanced research capabilities for generating more informed subtasks. If the Perplexity API is unavailable or encounters an error, the script will automatically fall back to using Anthropic's Claude.
-
-To use the Perplexity integration:
-
-1. Obtain a Perplexity API key
-2. Add `PERPLEXITY_API_KEY` to your `.env` file
-3. Optionally specify `PERPLEXITY_MODEL` in your `.env` file (default: "sonar-medium-online")
-4. Use the `--research` flag with the `expand` command
+- The script now uses a unified AI service layer (`ai-services-unified.js`).
+- Model selection (e.g., Claude vs. Perplexity for `--research`) is determined by the configuration in `.taskmasterconfig` based on the requested `role` (`main` or `research`).
+- API keys are automatically resolved from your `.env` file (for CLI) or MCP session environment.
+- To use the research capabilities (e.g., `expand --research`), ensure you have:
+  1.  Configured a model for the `research` role using `task-master models --setup` (Perplexity models are recommended).
+  2.  Added the corresponding API key (e.g., `PERPLEXITY_API_KEY`) to your `.env` file.
 
 ## Logging
 

docs/ai-client-utils-example.md

@@ -1,257 +0,0 @@
-# AI Client Utilities for MCP Tools
-
-This document provides examples of how to use the new AI client utilities with AsyncOperationManager in MCP tools.
-
-## Basic Usage with Direct Functions
-
-```javascript
-// In your direct function implementation:
-import {
-	getAnthropicClientForMCP,
-	getModelConfig,
-	handleClaudeError
-} from '../utils/ai-client-utils.js';
-
-export async function someAiOperationDirect(args, log, context) {
-	try {
-		// Initialize Anthropic client with session from context
-		const client = getAnthropicClientForMCP(context.session, log);
-
-		// Get model configuration with defaults or session overrides
-		const modelConfig = getModelConfig(context.session);
-
-		// Make API call with proper error handling
-		try {
-			const response = await client.messages.create({
-				model: modelConfig.model,
-				max_tokens: modelConfig.maxTokens,
-				temperature: modelConfig.temperature,
-				messages: [{ role: 'user', content: 'Your prompt here' }]
-			});
-
-			return {
-				success: true,
-				data: response
-			};
-		} catch (apiError) {
-			// Use helper to get user-friendly error message
-			const friendlyMessage = handleClaudeError(apiError);
-
-			return {
-				success: false,
-				error: {
-					code: 'AI_API_ERROR',
-					message: friendlyMessage
-				}
-			};
-		}
-	} catch (error) {
-		// Handle client initialization errors
-		return {
-			success: false,
-			error: {
-				code: 'AI_CLIENT_ERROR',
-				message: error.message
-			}
-		};
-	}
-}
-```
-
-## Integration with AsyncOperationManager
-
-```javascript
-// In your MCP tool implementation:
-import {
-	AsyncOperationManager,
-	StatusCodes
-} from '../../utils/async-operation-manager.js';
-import { someAiOperationDirect } from '../../core/direct-functions/some-ai-operation.js';
-
-export async function someAiOperation(args, context) {
-	const { session, mcpLog } = context;
-	const log = mcpLog || console;
-
-	try {
-		// Create operation description
-		const operationDescription = `AI operation: ${args.someParam}`;
-
-		// Start async operation
-		const operation = AsyncOperationManager.createOperation(
-			operationDescription,
-			async (reportProgress) => {
-				try {
-					// Initial progress report
-					reportProgress({
-						progress: 0,
-						status: 'Starting AI operation...'
-					});
-
-					// Call direct function with session and progress reporting
-					const result = await someAiOperationDirect(args, log, {
-						reportProgress,
-						mcpLog: log,
-						session
-					});
-
-					// Final progress update
-					reportProgress({
-						progress: 100,
-						status: result.success ? 'Operation completed' : 'Operation failed',
-						result: result.data,
-						error: result.error
-					});
-
-					return result;
-				} catch (error) {
-					// Handle errors in the operation
-					reportProgress({
-						progress: 100,
-						status: 'Operation failed',
-						error: {
-							message: error.message,
-							code: error.code || 'OPERATION_FAILED'
-						}
-					});
-					throw error;
-				}
-			}
-		);
-
-		// Return immediate response with operation ID
-		return {
-			status: StatusCodes.ACCEPTED,
-			body: {
-				success: true,
-				message: 'Operation started',
-				operationId: operation.id
-			}
-		};
-	} catch (error) {
-		// Handle errors in the MCP tool
-		log.error(`Error in someAiOperation: ${error.message}`);
-		return {
-			status: StatusCodes.INTERNAL_SERVER_ERROR,
-			body: {
-				success: false,
-				error: {
-					code: 'OPERATION_FAILED',
-					message: error.message
-				}
-			}
-		};
-	}
-}
-```
-
-## Using Research Capabilities with Perplexity
-
-```javascript
-// In your direct function:
-import {
-	getPerplexityClientForMCP,
-	getBestAvailableAIModel
-} from '../utils/ai-client-utils.js';
-
-export async function researchOperationDirect(args, log, context) {
-	try {
-		// Get the best AI model for this operation based on needs
-		const { type, client } = await getBestAvailableAIModel(
-			context.session,
-			{ requiresResearch: true },
-			log
-		);
-
-		// Report which model we're using
-		if (context.reportProgress) {
-			await context.reportProgress({
-				progress: 10,
-				status: `Using ${type} model for research...`
-			});
-		}
-
-		// Make API call based on the model type
-		if (type === 'perplexity') {
-			// Call Perplexity
-			const response = await client.chat.completions.create({
-				model: context.session?.env?.PERPLEXITY_MODEL || 'sonar-medium-online',
-				messages: [{ role: 'user', content: args.researchQuery }],
-				temperature: 0.1
-			});
-
-			return {
-				success: true,
-				data: response.choices[0].message.content
-			};
-		} else {
-			// Call Claude as fallback
-			// (Implementation depends on specific needs)
-			// ...
-		}
-	} catch (error) {
-		// Handle errors
-		return {
-			success: false,
-			error: {
-				code: 'RESEARCH_ERROR',
-				message: error.message
-			}
-		};
-	}
-}
-```
-
-## Model Configuration Override Example
-
-```javascript
-// In your direct function:
-import { getModelConfig } from '../utils/ai-client-utils.js';
-
-// Using custom defaults for a specific operation
-const operationDefaults = {
-	model: 'claude-3-haiku-20240307', // Faster, smaller model
-	maxTokens: 1000, // Lower token limit
-	temperature: 0.2 // Lower temperature for more deterministic output
-};
-
-// Get model config with operation-specific defaults
-const modelConfig = getModelConfig(context.session, operationDefaults);
-
-// Now use modelConfig in your API calls
-const response = await client.messages.create({
-	model: modelConfig.model,
-	max_tokens: modelConfig.maxTokens,
-	temperature: modelConfig.temperature
-	// Other parameters...
-});
-```
-
-## Best Practices
-
-1. **Error Handling**:
-
-   - Always use try/catch blocks around both client initialization and API calls
-   - Use `handleClaudeError` to provide user-friendly error messages
-   - Return standardized error objects with code and message
-
-2. **Progress Reporting**:
-
-   - Report progress at key points (starting, processing, completing)
-   - Include meaningful status messages
-   - Include error details in progress reports when failures occur
-
-3. **Session Handling**:
-
-   - Always pass the session from the context to the AI client getters
-   - Use `getModelConfig` to respect user settings from session
-
-4. **Model Selection**:
-
-   - Use `getBestAvailableAIModel` when you need to select between different models
-   - Set `requiresResearch: true` when you need Perplexity capabilities
-
-5. **AsyncOperationManager Integration**:
-   - Create descriptive operation names
-   - Handle all errors within the operation function
-   - Return standardized results from direct functions
-   - Return immediate responses with operation IDs

docs/command-reference.md

@@ -52,6 +52,9 @@ task-master show 1.2
 ```bash
 # Update tasks from a specific ID and provide context
 task-master update --from=<id> --prompt="<prompt>"
+
+# Update tasks using research role
+task-master update --from=<id> --prompt="<prompt>" --research
 ```
 
 ## Update a Specific Task
@@ -60,7 +63,7 @@ task-master update --from=<id> --prompt="<prompt>"
 # Update a single task by ID with new information
 task-master update-task --id=<id> --prompt="<prompt>"
 
-# Use research-backed updates with Perplexity AI
+# Use research-backed updates
 task-master update-task --id=<id> --prompt="<prompt>" --research
 ```
 
@@ -73,7 +76,7 @@ task-master update-subtask --id=<parentId.subtaskId> --prompt="<prompt>"
 # Example: Add details about API rate limiting to subtask 2 of task 5
 task-master update-subtask --id=5.2 --prompt="Add rate limiting of 100 requests per minute"
 
-# Use research-backed updates with Perplexity AI
+# Use research-backed updates
 task-master update-subtask --id=<parentId.subtaskId> --prompt="<prompt>" --research
 ```
 
@@ -187,9 +190,12 @@ task-master fix-dependencies
 ## Add a New Task
 
 ```bash
-# Add a new task using AI
+# Add a new task using AI (main role)
 task-master add-task --prompt="Description of the new task"
 
+# Add a new task using AI (research role)
+task-master add-task --prompt="Description of the new task" --research
+
 # Add a task with dependencies
 task-master add-task --prompt="Description" --dependencies=1,2,3
 
@@ -203,3 +209,30 @@ task-master add-task --prompt="Description" --priority=high
 # Initialize a new project with Task Master structure
 task-master init
 ```
+
+## Configure AI Models
+
+```bash
+# View current AI model configuration and API key status
+task-master models
+
+# Set the primary model for generation/updates (provider inferred if known)
+task-master models --set-main=claude-3-opus-20240229
+
+# Set the research model
+task-master models --set-research=sonar-pro
+
+# Set the fallback model
+task-master models --set-fallback=claude-3-haiku-20240307
+
+# Set a custom Ollama model for the main role
+task-master models --set-main=my-local-llama --ollama
+
+# Set a custom OpenRouter model for the research role
+task-master models --set-research=google/gemini-pro --openrouter
+
+# Run interactive setup to configure models, including custom ones
+task-master models --setup
+```
+
+Configuration is stored in `.taskmasterconfig` in your project root. API keys are still managed via `.env` or MCP configuration. Use `task-master models` without flags to see available built-in models. Use `--setup` for a guided experience.

docs/configuration.md

@@ -1,53 +1,89 @@
 # Configuration
 
-Task Master can be configured through environment variables in a `.env` file at the root of your project.
+Taskmaster uses two primary methods for configuration:
 
-## Required Configuration
+1.  **`.taskmasterconfig` File (Project Root - Recommended for most settings)**
 
-- `ANTHROPIC_API_KEY`: Your Anthropic API key for Claude (Example: `ANTHROPIC_API_KEY=sk-ant-api03-...`)
+    - This JSON file stores most configuration settings, including AI model selections, parameters, logging levels, and project defaults.
+    - **Location:** This file is created in the root directory of your project when you run the `task-master models --setup` interactive setup. You typically do this during the initialization sequence. Do not manually edit this file beyond adjusting Temperature and Max Tokens depending on your model.
+    - **Management:** Use the `task-master models --setup` command (or `models` MCP tool) to interactively create and manage this file. You can also set specific models directly using `task-master models --set-<role>=<model_id>`, adding `--ollama` or `--openrouter` flags for custom models. Manual editing is possible but not recommended unless you understand the structure.
+    - **Example Structure:**
+      ```json
+      {
+      	"models": {
+      		"main": {
+      			"provider": "anthropic",
+      			"modelId": "claude-3-7-sonnet-20250219",
+      			"maxTokens": 64000,
+      			"temperature": 0.2
+      		},
+      		"research": {
+      			"provider": "perplexity",
+      			"modelId": "sonar-pro",
+      			"maxTokens": 8700,
+      			"temperature": 0.1
+      		},
+      		"fallback": {
+      			"provider": "anthropic",
+      			"modelId": "claude-3-5-sonnet",
+      			"maxTokens": 64000,
+      			"temperature": 0.2
+      		}
+      	},
+      	"global": {
+      		"logLevel": "info",
+      		"debug": false,
+      		"defaultSubtasks": 5,
+      		"defaultPriority": "medium",
+      		"projectName": "Your Project Name",
+      		"ollamaBaseUrl": "http://localhost:11434/api",
+      		"azureOpenaiBaseUrl": "https://your-endpoint.openai.azure.com/"
+      	}
+      }
+      ```
 
-## Optional Configuration
+2.  **Environment Variables (`.env` file or MCP `env` block - For API Keys Only)**
+    - Used **exclusively** for sensitive API keys and specific endpoint URLs.
+    - **Location:**
+      - For CLI usage: Create a `.env` file in your project root.
+      - For MCP/Cursor usage: Configure keys in the `env` section of your `.cursor/mcp.json` file.
+    - **Required API Keys (Depending on configured providers):**
+      - `ANTHROPIC_API_KEY`: Your Anthropic API key.
+      - `PERPLEXITY_API_KEY`: Your Perplexity API key.
+      - `OPENAI_API_KEY`: Your OpenAI API key.
+      - `GOOGLE_API_KEY`: Your Google API key.
+      - `MISTRAL_API_KEY`: Your Mistral API key.
+      - `AZURE_OPENAI_API_KEY`: Your Azure OpenAI API key (also requires `AZURE_OPENAI_ENDPOINT`).
+      - `OPENROUTER_API_KEY`: Your OpenRouter API key.
+      - `XAI_API_KEY`: Your X-AI API key.
+    - **Optional Endpoint Overrides (in .taskmasterconfig):**
+      - `AZURE_OPENAI_ENDPOINT`: Required if using Azure OpenAI key.
+      - `OLLAMA_BASE_URL`: Override the default Ollama API URL (Default: `http://localhost:11434/api`).
 
-- `MODEL` (Default: `"claude-3-7-sonnet-20250219"`): Claude model to use (Example: `MODEL=claude-3-opus-20240229`)
-- `MAX_TOKENS` (Default: `"4000"`): Maximum tokens for responses (Example: `MAX_TOKENS=8000`)
-- `TEMPERATURE` (Default: `"0.7"`): Temperature for model responses (Example: `TEMPERATURE=0.5`)
-- `DEBUG` (Default: `"false"`): Enable debug logging (Example: `DEBUG=true`)
-- `LOG_LEVEL` (Default: `"info"`): Console output level (Example: `LOG_LEVEL=debug`)
-- `DEFAULT_SUBTASKS` (Default: `"3"`): Default subtask count (Example: `DEFAULT_SUBTASKS=5`)
-- `DEFAULT_PRIORITY` (Default: `"medium"`): Default priority (Example: `DEFAULT_PRIORITY=high`)
-- `PROJECT_NAME` (Default: `"MCP SaaS MVP"`): Project name in metadata (Example: `PROJECT_NAME=My Awesome Project`)
-- `PROJECT_VERSION` (Default: `"1.0.0"`): Version in metadata (Example: `PROJECT_VERSION=2.1.0`)
-- `PERPLEXITY_API_KEY`: For research-backed features (Example: `PERPLEXITY_API_KEY=pplx-...`)
-- `PERPLEXITY_MODEL` (Default: `"sonar-medium-online"`): Perplexity model (Example: `PERPLEXITY_MODEL=sonar-large-online`)
+**Important:** Settings like model ID selections (`main`, `research`, `fallback`), `maxTokens`, `temperature`, `logLevel`, `defaultSubtasks`, `defaultPriority`, and `projectName` are **managed in `.taskmasterconfig`**, not environment variables.
 
-## Example .env File
+## Example `.env` File (for API Keys)
 
 ```
-# Required
-ANTHROPIC_API_KEY=sk-ant-api03-your-api-key
+# Required API keys for providers configured in .taskmasterconfig
+ANTHROPIC_API_KEY=sk-ant-api03-your-key-here
+PERPLEXITY_API_KEY=pplx-your-key-here
+# OPENAI_API_KEY=sk-your-key-here
+# GOOGLE_API_KEY=AIzaSy...
+# etc.
 
-# Optional - Claude Configuration
-MODEL=claude-3-7-sonnet-20250219
-MAX_TOKENS=4000
-TEMPERATURE=0.7
-
-# Optional - Perplexity API for Research
-PERPLEXITY_API_KEY=pplx-your-api-key
-PERPLEXITY_MODEL=sonar-medium-online
-
-# Optional - Project Info
-PROJECT_NAME=My Project
-PROJECT_VERSION=1.0.0
-
-# Optional - Application Configuration
-DEFAULT_SUBTASKS=3
-DEFAULT_PRIORITY=medium
-DEBUG=false
-LOG_LEVEL=info
+# Optional Endpoint Overrides
+# AZURE_OPENAI_ENDPOINT=https://your-azure-endpoint.openai.azure.com/
+# OLLAMA_BASE_URL=http://custom-ollama-host:11434/api
 ```
 
 ## Troubleshooting
 
+### Configuration Errors
+
+- If Task Master reports errors about missing configuration or cannot find `.taskmasterconfig`, run `task-master models --setup` in your project root to create or repair the file.
+- Ensure API keys are correctly placed in your `.env` file (for CLI) or `.cursor/mcp.json` (for MCP) and are valid for the providers selected in `.taskmasterconfig`.
+
 ### If `task-master init` doesn't respond:
 
 Try running it with Node directly:

docs/contributor-docs/testing-roo-integration.md

@@ -0,0 +1,94 @@
+# Testing Roo Integration
+
+This document provides instructions for testing the Roo integration in the Task Master package.
+
+## Running Tests
+
+To run the tests for the Roo integration:
+
+```bash
+# Run all tests
+npm test
+
+# Run only Roo integration tests
+npm test -- -t "Roo"
+
+# Run specific test file
+npm test -- tests/integration/roo-files-inclusion.test.js
+```
+
+## Manual Testing
+
+To manually verify that the Roo files are properly included in the package:
+
+1. Create a test directory:
+
+   ```bash
+   mkdir test-tm
+   cd test-tm
+   ```
+
+2. Create a package.json file:
+
+   ```bash
+   npm init -y
+   ```
+
+3. Install the task-master-ai package locally:
+
+   ```bash
+   # From the root of the claude-task-master repository
+   cd ..
+   npm pack
+   # This will create a file like task-master-ai-0.12.0.tgz
+
+   # Move back to the test directory
+   cd test-tm
+   npm install ../task-master-ai-0.12.0.tgz
+   ```
+
+4. Initialize a new Task Master project:
+
+   ```bash
+   npx task-master init --yes
+   ```
+
+5. Verify that all Roo files and directories are created:
+
+   ```bash
+   # Check that .roomodes file exists
+   ls -la | grep .roomodes
+
+   # Check that .roo directory exists and contains all mode directories
+   ls -la .roo
+   ls -la .roo/rules
+   ls -la .roo/rules-architect
+   ls -la .roo/rules-ask
+   ls -la .roo/rules-boomerang
+   ls -la .roo/rules-code
+   ls -la .roo/rules-debug
+   ls -la .roo/rules-test
+   ```
+
+## What to Look For
+
+When running the tests or performing manual verification, ensure that:
+
+1. The package includes `.roo/**` and `.roomodes` in the `files` array in package.json
+2. The `prepare-package.js` script verifies the existence of all required Roo files
+3. The `init.js` script creates all necessary .roo directories and copies .roomodes file
+4. All source files for Roo integration exist in `assets/roocode/.roo` and `assets/roocode/.roomodes`
+
+## Compatibility
+
+Ensure that the Roo integration works alongside existing Cursor functionality:
+
+1. Initialize a new project that uses both Cursor and Roo:
+
+   ```bash
+   npx task-master init --yes
+   ```
+
+2. Verify that both `.cursor` and `.roo` directories are created
+3. Verify that both `.windsurfrules` and `.roomodes` files are created
+4. Confirm that existing functionality continues to work as expected

docs/examples.md

@@ -51,3 +51,33 @@ Can you analyze the complexity of our tasks to help me understand which ones nee
 ```
 Can you show me the complexity report in a more readable format?
 ```
+
+### Breaking Down Complex Tasks
+
+```
+Task 5 seems complex. Can you break it down into subtasks?
+```
+
+(Agent runs: `task-master expand --id=5`)
+
+```
+Please break down task 5 using research-backed generation.
+```
+
+(Agent runs: `task-master expand --id=5 --research`)
+
+### Updating Tasks with Research
+
+```
+We need to update task 15 based on the latest React Query v5 changes. Can you research this and update the task?
+```
+
+(Agent runs: `task-master update-task --id=15 --prompt="Update based on React Query v5 changes" --research`)
+
+### Adding Tasks with Research
+
+```
+Please add a new task to implement user profile image uploads using Cloudinary, research the best approach.
+```
+
+(Agent runs: `task-master add-task --prompt="Implement user profile image uploads using Cloudinary" --research`)

docs/tutorial.md

@@ -10,32 +10,45 @@ There are two ways to set up Task Master: using MCP (recommended) or via npm ins
 
 MCP (Model Control Protocol) provides the easiest way to get started with Task Master directly in your editor.
 
-1. **Add the MCP config to your editor** (Cursor recommended, but it works with other text editors):
+1. **Install the package**
+
+```bash
+npm i -g task-master-ai
+```
+
+2. **Add the MCP config to your IDE/MCP Client** (Cursor is recommended, but it works with other clients):
 
 ```json
 {
 	"mcpServers": {
 		"taskmaster-ai": {
 			"command": "npx",
-			"args": ["-y", "task-master-ai"],
+			"args": ["-y", "--package=task-master-ai", "task-master-ai"],
 			"env": {
 				"ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
 				"PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY_HERE",
-				"MODEL": "claude-3-7-sonnet-20250219",
-				"PERPLEXITY_MODEL": "sonar-pro",
-				"MAX_TOKENS": 64000,
-				"TEMPERATURE": 0.2,
-				"DEFAULT_SUBTASKS": 5,
-				"DEFAULT_PRIORITY": "medium"
+				"OPENAI_API_KEY": "YOUR_OPENAI_KEY_HERE",
+				"GOOGLE_API_KEY": "YOUR_GOOGLE_KEY_HERE",
+				"MISTRAL_API_KEY": "YOUR_MISTRAL_KEY_HERE",
+				"OPENROUTER_API_KEY": "YOUR_OPENROUTER_KEY_HERE",
+				"XAI_API_KEY": "YOUR_XAI_KEY_HERE",
+				"AZURE_OPENAI_API_KEY": "YOUR_AZURE_KEY_HERE"
 			}
 		}
 	}
 }
 ```
 
-2. **Enable the MCP** in your editor settings
+**IMPORTANT:** An API key is _required_ for each AI provider you plan on using. Run the `task-master models` command to see your selected models and the status of your API keys across .env and mcp.json
 
-3. **Prompt the AI** to initialize Task Master:
+**To use AI commands in CLI** you MUST have API keys in the .env file
+**To use AI commands in MCP** you MUST have API keys in the .mcp.json file (or MCP config equivalent)
+
+We recommend having keys in both places and adding mcp.json to your gitignore so your API keys aren't checked into git.
+
+3. **Enable the MCP** in your editor settings
+
+4. **Prompt the AI** to initialize Task Master:
 
 ```
 Can you please initialize taskmaster-ai into my project?
@@ -47,9 +60,9 @@ The AI will:
 - Set up initial configuration files
 - Guide you through the rest of the process
 
-4. Place your PRD document in the `scripts/` directory (e.g., `scripts/prd.txt`)
+5. Place your PRD document in the `scripts/` directory (e.g., `scripts/prd.txt`)
 
-5. **Use natural language commands** to interact with Task Master:
+6. **Use natural language commands** to interact with Task Master:
 
 ```
 Can you parse my PRD at scripts/prd.txt?
@@ -76,7 +89,7 @@ Initialize a new project:
 task-master init
 
 # If installed locally
-npx task-master-init
+npx task-master init
 ```
 
 This will prompt you for project details and set up a new project with the necessary files and structure.
@@ -132,7 +145,7 @@ You can also set up the MCP server in Cursor settings:
 4. Configure with the following details:
    - Name: "Task Master"
    - Type: "Command"
-   - Command: "npx -y task-master-mcp"
+   - Command: "npx -y --package=task-master-ai task-master-ai"
 5. Save the settings
 
 Once configured, you can interact with Task Master's task management commands directly through Cursor's interface, providing a more integrated experience.
@@ -241,13 +254,16 @@ If during implementation, you discover that:
 Tell the agent:
 
 ```
-We've changed our approach. We're now using Express instead of Fastify. Please update all future tasks to reflect this change.
+We've decided to use MongoDB instead of PostgreSQL. Can you update all future tasks (from ID 4) to reflect this change?
 ```
 
 The agent will execute:
 
 ```bash
-task-master update --from=4 --prompt="Now we are using Express instead of Fastify."
+task-master update --from=4 --prompt="Now we are using MongoDB instead of PostgreSQL."
+
+# OR, if research is needed to find best practices for MongoDB:
+task-master update --from=4 --prompt="Update to use MongoDB, researching best practices" --research
 ```
 
 This will rewrite or re-scope subsequent tasks in tasks.json while preserving completed work.
@@ -290,7 +306,7 @@ The agent will execute:
 task-master expand --all
 ```
 
-For research-backed subtask generation using Perplexity AI:
+For research-backed subtask generation using the configured research model:
 
 ```
 Please break down task 5 using research-backed generation.

index.js

@@ -46,22 +46,18 @@ export const initProject = async (options = {}) => {
 };
 
 // Export a function to run init as a CLI command
-export const runInitCLI = async () => {
-	// Using spawn to ensure proper handling of stdio and process exit
-	const child = spawn('node', [resolve(__dirname, './scripts/init.js')], {
-		stdio: 'inherit',
-		cwd: process.cwd()
-	});
-
-	return new Promise((resolve, reject) => {
-		child.on('close', (code) => {
-			if (code === 0) {
-				resolve();
-			} else {
-				reject(new Error(`Init script exited with code ${code}`));
-			}
-		});
-	});
+export const runInitCLI = async (options = {}) => {
+	try {
+		const init = await import('./scripts/init.js');
+		const result = await init.initializeProject(options);
+		return result;
+	} catch (error) {
+		console.error('Initialization failed:', error.message);
+		if (process.env.DEBUG === 'true') {
+			console.error('Debug stack trace:', error.stack);
+		}
+		throw error; // Re-throw to be handled by the command handler
+	}
 };
 
 // Export version information
@@ -79,11 +75,21 @@ if (import.meta.url === `file://${process.argv[1]}`) {
 	program
 		.command('init')
 		.description('Initialize a new project')
-		.action(() => {
-			runInitCLI().catch((err) => {
+		.option('-y, --yes', 'Skip prompts and use default values')
+		.option('-n, --name <n>', 'Project name')
+		.option('-d, --description <description>', 'Project description')
+		.option('-v, --version <version>', 'Project version', '0.1.0')
+		.option('-a, --author <author>', 'Author name')
+		.option('--skip-install', 'Skip installing dependencies')
+		.option('--dry-run', 'Show what would be done without making changes')
+		.option('--aliases', 'Add shell aliases (tm, taskmaster)')
+		.action(async (cmdOptions) => {
+			try {
+				await runInitCLI(cmdOptions);
+			} catch (err) {
 				console.error('Init failed:', err.message);
 				process.exit(1);
-			});
+			}
 		});
 
 	program

jest.config.js

@@ -15,11 +15,7 @@ export default {
 	roots: ['<rootDir>/tests'],
 
 	// The glob patterns Jest uses to detect test files
-	testMatch: [
-		'**/__tests__/**/*.js',
-		'**/?(*.)+(spec|test).js',
-		'**/tests/*.test.js'
-	],
+	testMatch: ['**/__tests__/**/*.js', '**/?(*.)+(spec|test).js'],
 
 	// Transform files
 	transform: {},

mcp-server/src/core/direct-functions/add-task.js

@@ -8,15 +8,7 @@ import {
 	enableSilentMode,
 	disableSilentMode
 } from '../../../../scripts/modules/utils.js';
-import {
-	getAnthropicClientForMCP,
-	getModelConfig
-} from '../utils/ai-client-utils.js';
-import {
-	_buildAddTaskPrompt,
-	parseTaskJsonResponse,
-	_handleAnthropicStream
-} from '../../../../scripts/modules/ai-services.js';
+import { createLogWrapper } from '../../tools/utils.js';
 
 /**
  * Direct function wrapper for adding a new task with error handling.
@@ -29,20 +21,32 @@ import {
  * @param {string} [args.testStrategy] - Test strategy (for manual task creation)
  * @param {string} [args.dependencies] - Comma-separated list of task IDs this task depends on
  * @param {string} [args.priority='medium'] - Task priority (high, medium, low)
- * @param {string} [args.file='tasks/tasks.json'] - Path to the tasks file
- * @param {string} [args.projectRoot] - Project root directory
+ * @param {string} [args.tasksJsonPath] - Path to the tasks.json file (resolved by tool)
  * @param {boolean} [args.research=false] - Whether to use research capabilities for task creation
+ * @param {string} [args.projectRoot] - Project root path
  * @param {Object} log - Logger object
- * @param {Object} context - Additional context (reportProgress, session)
+ * @param {Object} context - Additional context (session)
  * @returns {Promise<Object>} - Result object { success: boolean, data?: any, error?: { code: string, message: string } }
  */
 export async function addTaskDirect(args, log, context = {}) {
-	// Destructure expected args
-	const { tasksJsonPath, prompt, dependencies, priority, research } = args;
-	try {
-		// Enable silent mode to prevent console logs from interfering with JSON response
-		enableSilentMode();
+	// Destructure expected args (including research and projectRoot)
+	const {
+		tasksJsonPath,
+		prompt,
+		dependencies,
+		priority,
+		research,
+		projectRoot
+	} = args;
+	const { session } = context; // Destructure session from context
 
+	// Enable silent mode to prevent console logs from interfering with JSON response
+	enableSilentMode();
+
+	// Create logger wrapper using the utility
+	const mcpLog = createLogWrapper(log);
+
+	try {
 		// Check if tasksJsonPath was provided
 		if (!tasksJsonPath) {
 			log.error('addTaskDirect called without tasksJsonPath');
@@ -79,20 +83,17 @@ export async function addTaskDirect(args, log, context = {}) {
 		}
 
 		// Extract and prepare parameters
-		const taskPrompt = prompt;
 		const taskDependencies = Array.isArray(dependencies)
-			? dependencies
-			: dependencies
+			? dependencies // Already an array if passed directly
+			: dependencies // Check if dependencies exist and are a string
 				? String(dependencies)
 						.split(',')
-						.map((id) => parseInt(id.trim(), 10))
-				: [];
-		const taskPriority = priority || 'medium';
-
-		// Extract context parameters for advanced functionality
-		const { session } = context;
+						.map((id) => parseInt(id.trim(), 10)) // Split, trim, and parse
+				: []; // Default to empty array if null/undefined
+		const taskPriority = priority || 'medium'; // Default priority
 
 		let manualTaskData = null;
+		let newTaskId;
 
 		if (isManualCreation) {
 			// Create manual task data object
@@ -108,150 +109,64 @@ export async function addTaskDirect(args, log, context = {}) {
 			);
 
 			// Call the addTask function with manual task data
-			const newTaskId = await addTask(
+			newTaskId = await addTask(
 				tasksPath,
-				null, // No prompt needed for manual creation
+				null, // prompt is null for manual creation
 				taskDependencies,
-				priority,
+				taskPriority,
 				{
-					mcpLog: log,
-					session
+					session,
+					mcpLog,
+					projectRoot
 				},
-				'json', // Use JSON output format to prevent console output
-				null, // No custom environment
-				manualTaskData // Pass the manual task data
+				'json', // outputFormat
+				manualTaskData, // Pass the manual task data
+				false, // research flag is false for manual creation
+				projectRoot // Pass projectRoot
 			);
-
-			// Restore normal logging
-			disableSilentMode();
-
-			return {
-				success: true,
-				data: {
-					taskId: newTaskId,
-					message: `Successfully added new task #${newTaskId}`
-				}
-			};
 		} else {
 			// AI-driven task creation
 			log.info(
-				`Adding new task with prompt: "${prompt}", dependencies: [${taskDependencies.join(', ')}], priority: ${priority}`
+				`Adding new task with prompt: "${prompt}", dependencies: [${taskDependencies.join(', ')}], priority: ${taskPriority}, research: ${research}`
 			);
 
-			// Initialize AI client with session environment
-			let localAnthropic;
-			try {
-				localAnthropic = getAnthropicClientForMCP(session, log);
-			} catch (error) {
-				log.error(`Failed to initialize Anthropic client: ${error.message}`);
-				disableSilentMode();
-				return {
-					success: false,
-					error: {
-						code: 'AI_CLIENT_ERROR',
-						message: `Cannot initialize AI client: ${error.message}`
-					}
-				};
-			}
-
-			// Get model configuration from session
-			const modelConfig = getModelConfig(session);
-
-			// Read existing tasks to provide context
-			let tasksData;
-			try {
-				const fs = await import('fs');
-				tasksData = JSON.parse(fs.readFileSync(tasksPath, 'utf8'));
-			} catch (error) {
-				log.warn(`Could not read existing tasks for context: ${error.message}`);
-				tasksData = { tasks: [] };
-			}
-
-			// Build prompts for AI
-			const { systemPrompt, userPrompt } = _buildAddTaskPrompt(
-				prompt,
-				tasksData.tasks
-			);
-
-			// Make the AI call using the streaming helper
-			let responseText;
-			try {
-				responseText = await _handleAnthropicStream(
-					localAnthropic,
-					{
-						model: modelConfig.model,
-						max_tokens: modelConfig.maxTokens,
-						temperature: modelConfig.temperature,
-						messages: [{ role: 'user', content: userPrompt }],
-						system: systemPrompt
-					},
-					{
-						mcpLog: log
-					}
-				);
-			} catch (error) {
-				log.error(`AI processing failed: ${error.message}`);
-				disableSilentMode();
-				return {
-					success: false,
-					error: {
-						code: 'AI_PROCESSING_ERROR',
-						message: `Failed to generate task with AI: ${error.message}`
-					}
-				};
-			}
-
-			// Parse the AI response
-			let taskDataFromAI;
-			try {
-				taskDataFromAI = parseTaskJsonResponse(responseText);
-			} catch (error) {
-				log.error(`Failed to parse AI response: ${error.message}`);
-				disableSilentMode();
-				return {
-					success: false,
-					error: {
-						code: 'RESPONSE_PARSING_ERROR',
-						message: `Failed to parse AI response: ${error.message}`
-					}
-				};
-			}
-
-			// Call the addTask function with 'json' outputFormat to prevent console output when called via MCP
-			const newTaskId = await addTask(
+			// Call the addTask function, passing the research flag
+			newTaskId = await addTask(
 				tasksPath,
-				prompt,
+				prompt, // Use the prompt for AI creation
 				taskDependencies,
-				priority,
+				taskPriority,
 				{
-					mcpLog: log,
-					session
+					session,
+					mcpLog,
+					projectRoot
 				},
-				'json',
-				null,
-				taskDataFromAI // Pass the parsed AI result as the manual task data
+				'json', // outputFormat
+				null, // manualTaskData is null for AI creation
+				research // Pass the research flag
 			);
-
-			// Restore normal logging
-			disableSilentMode();
-
-			return {
-				success: true,
-				data: {
-					taskId: newTaskId,
-					message: `Successfully added new task #${newTaskId}`
-				}
-			};
 		}
+
+		// Restore normal logging
+		disableSilentMode();
+
+		return {
+			success: true,
+			data: {
+				taskId: newTaskId,
+				message: `Successfully added new task #${newTaskId}`
+			}
+		};
 	} catch (error) {
 		// Make sure to restore normal logging even if there's an error
 		disableSilentMode();
 
 		log.error(`Error in addTaskDirect: ${error.message}`);
+		// Add specific error code checks if needed
 		return {
 			success: false,
 			error: {
-				code: 'ADD_TASK_ERROR',
+				code: error.code || 'ADD_TASK_ERROR', // Use error code if available
 				message: error.message
 			}
 		};

mcp-server/src/core/direct-functions/analyze-task-complexity.js

@@ -2,37 +2,38 @@
  * Direct function wrapper for analyzeTaskComplexity
  */
 
-import { analyzeTaskComplexity } from '../../../../scripts/modules/task-manager.js';
+import analyzeTaskComplexity from '../../../../scripts/modules/task-manager/analyze-task-complexity.js';
 import {
 	enableSilentMode,
 	disableSilentMode,
-	isSilentMode,
-	readJSON
+	isSilentMode
 } from '../../../../scripts/modules/utils.js';
 import fs from 'fs';
-import path from 'path';
+import { createLogWrapper } from '../../tools/utils.js'; // Import the new utility
 
 /**
  * Analyze task complexity and generate recommendations
  * @param {Object} args - Function arguments
  * @param {string} args.tasksJsonPath - Explicit path to the tasks.json file.
  * @param {string} args.outputPath - Explicit absolute path to save the report.
- * @param {string} [args.model] - LLM model to use for analysis
  * @param {string|number} [args.threshold] - Minimum complexity score to recommend expansion (1-10)
  * @param {boolean} [args.research] - Use Perplexity AI for research-backed complexity analysis
+ * @param {string} [args.projectRoot] - Project root path.
  * @param {Object} log - Logger object
  * @param {Object} [context={}] - Context object containing session data
+ * @param {Object} [context.session] - MCP session object
  * @returns {Promise<{success: boolean, data?: Object, error?: {code: string, message: string}}>}
  */
 export async function analyzeTaskComplexityDirect(args, log, context = {}) {
-	const { session } = context; // Only extract session, not reportProgress
-	// Destructure expected args
-	const { tasksJsonPath, outputPath, model, threshold, research } = args;
+	const { session } = context;
+	const { tasksJsonPath, outputPath, threshold, research, projectRoot } = args;
 
+	const logWrapper = createLogWrapper(log);
+
+	// --- Initial Checks (remain the same) ---
 	try {
 		log.info(`Analyzing task complexity with args: ${JSON.stringify(args)}`);
 
-		// Check if required paths were provided
 		if (!tasksJsonPath) {
 			log.error('analyzeTaskComplexityDirect called without tasksJsonPath');
 			return {
@@ -51,7 +52,6 @@ export async function analyzeTaskComplexityDirect(args, log, context = {}) {
 			};
 		}
 
-		// Use the provided paths
 		const tasksPath = tasksJsonPath;
 		const resolvedOutputPath = outputPath;
 
@@ -59,78 +59,93 @@ export async function analyzeTaskComplexityDirect(args, log, context = {}) {
 		log.info(`Output report will be saved to: ${resolvedOutputPath}`);
 
 		if (research) {
-			log.info('Using Perplexity AI for research-backed complexity analysis');
+			log.info('Using research role for complexity analysis');
 		}
 
-		// Create options object for analyzeTaskComplexity using provided paths
-		const options = {
-			file: tasksPath,
-			output: resolvedOutputPath,
-			model: model,
+		// Prepare options for the core function - REMOVED mcpLog and session here
+		const coreOptions = {
+			file: tasksJsonPath,
+			output: outputPath,
 			threshold: threshold,
-			research: research === true
+			research: research === true, // Ensure boolean
+			projectRoot: projectRoot // Pass projectRoot here
 		};
+		// --- End Initial Checks ---
 
-		// Enable silent mode to prevent console logs from interfering with JSON response
+		// --- Silent Mode and Logger Wrapper ---
 		const wasSilent = isSilentMode();
 		if (!wasSilent) {
-			enableSilentMode();
+			enableSilentMode(); // Still enable silent mode as a backup
 		}
 
-		// Create a logWrapper that matches the expected mcpLog interface as specified in utilities.mdc
-		const logWrapper = {
-			info: (message, ...args) => log.info(message, ...args),
-			warn: (message, ...args) => log.warn(message, ...args),
-			error: (message, ...args) => log.error(message, ...args),
-			debug: (message, ...args) => log.debug && log.debug(message, ...args),
-			success: (message, ...args) => log.info(message, ...args) // Map success to info
-		};
+		let report;
 
 		try {
-			// Call the core function with session and logWrapper as mcpLog
-			await analyzeTaskComplexity(options, {
-				session,
-				mcpLog: logWrapper // Use the wrapper instead of passing log directly
-			});
+			// --- Call Core Function (Pass context separately) ---
+			// Pass coreOptions as the first argument
+			// Pass context object { session, mcpLog } as the second argument
+			report = await analyzeTaskComplexity(
+				coreOptions, // Pass options object
+				{ session, mcpLog: logWrapper } // Pass context object
+				// Removed the explicit 'json' format argument, assuming context handling is sufficient
+				// If issues persist, we might need to add an explicit format param to analyzeTaskComplexity
+			);
 		} catch (error) {
-			log.error(`Error in analyzeTaskComplexity: ${error.message}`);
+			log.error(
+				`Error in analyzeTaskComplexity core function: ${error.message}`
+			);
+			// Restore logging if we changed it
+			if (!wasSilent && isSilentMode()) {
+				disableSilentMode();
+			}
 			return {
 				success: false,
 				error: {
-					code: 'ANALYZE_ERROR',
-					message: `Error running complexity analysis: ${error.message}`
+					code: 'ANALYZE_CORE_ERROR',
+					message: `Error running core complexity analysis: ${error.message}`
 				}
 			};
 		} finally {
-			// Always restore normal logging in finally block, but only if we enabled it
-			if (!wasSilent) {
+			// Always restore normal logging in finally block if we enabled silent mode
+			if (!wasSilent && isSilentMode()) {
 				disableSilentMode();
 			}
 		}
 
-		// Verify the report file was created
+		// --- Result Handling (remains largely the same) ---
+		// Verify the report file was created (core function writes it)
 		if (!fs.existsSync(resolvedOutputPath)) {
 			return {
 				success: false,
 				error: {
-					code: 'ANALYZE_ERROR',
-					message: 'Analysis completed but no report file was created'
+					code: 'ANALYZE_REPORT_MISSING', // Specific code
+					message:
+						'Analysis completed but no report file was created at the expected path.'
+				}
+			};
+		}
+
+		// Added a check to ensure report is defined before accessing its properties
+		if (!report || typeof report !== 'object') {
+			log.error(
+				'Core analysis function returned an invalid or undefined response.'
+			);
+			return {
+				success: false,
+				error: {
+					code: 'INVALID_CORE_RESPONSE',
+					message: 'Core analysis function returned an invalid response.'
 				}
 			};
 		}
 
-		// Read the report file
-		let report;
 		try {
-			report = JSON.parse(fs.readFileSync(resolvedOutputPath, 'utf8'));
+			// Ensure complexityAnalysis exists and is an array
+			const analysisArray = Array.isArray(report.complexityAnalysis)
+				? report.complexityAnalysis
+				: [];
 
-			// Important: Handle different report formats
-			// The core function might return an array or an object with a complexityAnalysis property
-			const analysisArray = Array.isArray(report)
-				? report
-				: report.complexityAnalysis || [];
-
-			// Count tasks by complexity
+			// Count tasks by complexity (remains the same)
 			const highComplexityTasks = analysisArray.filter(
 				(t) => t.complexityScore >= 8
 			).length;
@@ -144,37 +159,40 @@ export async function analyzeTaskComplexityDirect(args, log, context = {}) {
 			return {
 				success: true,
 				data: {
-					message: `Task complexity analysis complete. Report saved to ${resolvedOutputPath}`,
-					reportPath: resolvedOutputPath,
+					message: `Task complexity analysis complete. Report saved to ${outputPath}`, // Use outputPath from args
+					reportPath: outputPath, // Use outputPath from args
 					reportSummary: {
 						taskCount: analysisArray.length,
 						highComplexityTasks,
 						mediumComplexityTasks,
 						lowComplexityTasks
-					}
+					},
+					fullReport: report // Now includes the full report
 				}
 			};
 		} catch (parseError) {
-			log.error(`Error parsing report file: ${parseError.message}`);
+			// Should not happen if core function returns object, but good safety check
+			log.error(`Internal error processing report data: ${parseError.message}`);
 			return {
 				success: false,
 				error: {
-					code: 'REPORT_PARSE_ERROR',
-					message: `Error parsing complexity report: ${parseError.message}`
+					code: 'REPORT_PROCESS_ERROR',
+					message: `Internal error processing complexity report: ${parseError.message}`
 				}
 			};
 		}
+		// --- End Result Handling ---
 	} catch (error) {
-		// Make sure to restore normal logging even if there's an error
+		// Catch errors from initial checks or path resolution
+		// Make sure to restore normal logging if silent mode was enabled
 		if (isSilentMode()) {
 			disableSilentMode();
 		}
-
-		log.error(`Error in analyzeTaskComplexityDirect: ${error.message}`);
+		log.error(`Error in analyzeTaskComplexityDirect setup: ${error.message}`);
 		return {
 			success: false,
 			error: {
-				code: 'CORE_FUNCTION_ERROR',
+				code: 'DIRECT_FUNCTION_SETUP_ERROR',
 				message: error.message
 			}
 		};

mcp-server/src/core/direct-functions/complexity-report.js

@@ -9,7 +9,6 @@ import {
 	disableSilentMode
 } from '../../../../scripts/modules/utils.js';
 import { getCachedOrExecute } from '../../tools/utils.js';
-import path from 'path';
 
 /**
  * Direct function wrapper for displaying the complexity report with error handling and caching.

mcp-server/src/core/direct-functions/expand-all-tasks.js

@@ -5,138 +5,86 @@
 import { expandAllTasks } from '../../../../scripts/modules/task-manager.js';
 import {
 	enableSilentMode,
-	disableSilentMode,
-	isSilentMode
+	disableSilentMode
 } from '../../../../scripts/modules/utils.js';
-import { getAnthropicClientForMCP } from '../utils/ai-client-utils.js';
-import path from 'path';
-import fs from 'fs';
+import { createLogWrapper } from '../../tools/utils.js';
 
 /**
- * Expand all pending tasks with subtasks
+ * Expand all pending tasks with subtasks (Direct Function Wrapper)
  * @param {Object} args - Function arguments
  * @param {string} args.tasksJsonPath - Explicit path to the tasks.json file.
  * @param {number|string} [args.num] - Number of subtasks to generate
- * @param {boolean} [args.research] - Enable Perplexity AI for research-backed subtask generation
+ * @param {boolean} [args.research] - Enable research-backed subtask generation
  * @param {string} [args.prompt] - Additional context to guide subtask generation
  * @param {boolean} [args.force] - Force regeneration of subtasks for tasks that already have them
- * @param {Object} log - Logger object
+ * @param {string} [args.projectRoot] - Project root path.
+ * @param {Object} log - Logger object from FastMCP
  * @param {Object} context - Context object containing session
  * @returns {Promise<{success: boolean, data?: Object, error?: {code: string, message: string}}>}
  */
 export async function expandAllTasksDirect(args, log, context = {}) {
-	const { session } = context; // Only extract session, not reportProgress
-	// Destructure expected args
-	const { tasksJsonPath, num, research, prompt, force } = args;
+	const { session } = context; // Extract session
+	// Destructure expected args, including projectRoot
+	const { tasksJsonPath, num, research, prompt, force, projectRoot } = args;
 
-	try {
-		log.info(`Expanding all tasks with args: ${JSON.stringify(args)}`);
+	// Create logger wrapper using the utility
+	const mcpLog = createLogWrapper(log);
 
-		// Check if tasksJsonPath was provided
-		if (!tasksJsonPath) {
-			log.error('expandAllTasksDirect called without tasksJsonPath');
-			return {
-				success: false,
-				error: {
-					code: 'MISSING_ARGUMENT',
-					message: 'tasksJsonPath is required'
-				}
-			};
-		}
-
-		// Enable silent mode early to prevent any console output
-		enableSilentMode();
-
-		try {
-			// Remove internal path finding
-			/*
-			const tasksPath = findTasksJsonPath(args, log);
-			*/
-			// Use provided path
-			const tasksPath = tasksJsonPath;
-
-			// Parse parameters
-			const numSubtasks = num ? parseInt(num, 10) : undefined;
-			const useResearch = research === true;
-			const additionalContext = prompt || '';
-			const forceFlag = force === true;
-
-			log.info(
-				`Expanding all tasks with ${numSubtasks || 'default'} subtasks each...`
-			);
-
-			if (useResearch) {
-				log.info('Using Perplexity AI for research-backed subtask generation');
-
-				// Initialize AI client for research-backed expansion
-				try {
-					await getAnthropicClientForMCP(session, log);
-				} catch (error) {
-					// Ensure silent mode is disabled before returning error
-					disableSilentMode();
-
-					log.error(`Failed to initialize AI client: ${error.message}`);
-					return {
-						success: false,
-						error: {
-							code: 'AI_CLIENT_ERROR',
-							message: `Cannot initialize AI client: ${error.message}`
-						}
-					};
-				}
-			}
-
-			if (additionalContext) {
-				log.info(`Additional context: "${additionalContext}"`);
-			}
-			if (forceFlag) {
-				log.info('Force regeneration of subtasks is enabled');
-			}
-
-			// Call the core function with session context for AI operations
-			// and outputFormat as 'json' to prevent UI elements
-			const result = await expandAllTasks(
-				tasksPath,
-				numSubtasks,
-				useResearch,
-				additionalContext,
-				forceFlag,
-				{ mcpLog: log, session },
-				'json' // Use JSON output format to prevent UI elements
-			);
-
-			// The expandAllTasks function now returns a result object
-			return {
-				success: true,
-				data: {
-					message: 'Successfully expanded all pending tasks with subtasks',
-					details: {
-						numSubtasks: numSubtasks,
-						research: useResearch,
-						prompt: additionalContext,
-						force: forceFlag,
-						tasksExpanded: result.expandedCount,
-						totalEligibleTasks: result.tasksToExpand
-					}
-				}
-			};
-		} finally {
-			// Restore normal logging in finally block to ensure it runs even if there's an error
-			disableSilentMode();
-		}
-	} catch (error) {
-		// Ensure silent mode is disabled if an error occurs
-		if (isSilentMode()) {
-			disableSilentMode();
-		}
-
-		log.error(`Error in expandAllTasksDirect: ${error.message}`);
+	if (!tasksJsonPath) {
+		log.error('expandAllTasksDirect called without tasksJsonPath');
 		return {
 			success: false,
 			error: {
-				code: 'CORE_FUNCTION_ERROR',
-				message: error.message
+				code: 'MISSING_ARGUMENT',
+				message: 'tasksJsonPath is required'
 			}
 		};
 	}
+
+	enableSilentMode(); // Enable silent mode for the core function call
+	try {
+		log.info(
+			`Calling core expandAllTasks with args: ${JSON.stringify({ num, research, prompt, force, projectRoot })}`
+		);
+
+		// Parse parameters (ensure correct types)
+		const numSubtasks = num ? parseInt(num, 10) : undefined;
+		const useResearch = research === true;
+		const additionalContext = prompt || '';
+		const forceFlag = force === true;
+
+		// Call the core function, passing options and the context object { session, mcpLog, projectRoot }
+		const result = await expandAllTasks(
+			tasksJsonPath,
+			numSubtasks,
+			useResearch,
+			additionalContext,
+			forceFlag,
+			{ session, mcpLog, projectRoot }
+		);
+
+		// Core function now returns a summary object
+		return {
+			success: true,
+			data: {
+				message: `Expand all operation completed. Expanded: ${result.expandedCount}, Failed: ${result.failedCount}, Skipped: ${result.skippedCount}`,
+				details: result // Include the full result details
+			}
+		};
+	} catch (error) {
+		// Log the error using the MCP logger
+		log.error(`Error during core expandAllTasks execution: ${error.message}`);
+		// Optionally log stack trace if available and debug enabled
+		// if (error.stack && log.debug) { log.debug(error.stack); }
+
+		return {
+			success: false,
+			error: {
+				code: 'CORE_FUNCTION_ERROR', // Or a more specific code if possible
+				message: error.message
+			}
+		};
+	} finally {
+		disableSilentMode(); // IMPORTANT: Ensure silent mode is always disabled
+	}
 }

mcp-server/src/core/direct-functions/expand-task.js

@@ -3,7 +3,7 @@
  * Direct function implementation for expanding a task into subtasks
  */
 
-import { expandTask } from '../../../../scripts/modules/task-manager.js';
+import expandTask from '../../../../scripts/modules/task-manager/expand-task.js';
 import {
 	readJSON,
 	writeJSON,
@@ -11,12 +11,9 @@ import {
 	disableSilentMode,
 	isSilentMode
 } from '../../../../scripts/modules/utils.js';
-import {
-	getAnthropicClientForMCP,
-	getModelConfig
-} from '../utils/ai-client-utils.js';
 import path from 'path';
 import fs from 'fs';
+import { createLogWrapper } from '../../tools/utils.js';
 
 /**
  * Direct function wrapper for expanding a task into subtasks with error handling.
@@ -25,17 +22,19 @@ import fs from 'fs';
  * @param {string} args.tasksJsonPath - Explicit path to the tasks.json file.
  * @param {string} args.id - The ID of the task to expand.
  * @param {number|string} [args.num] - Number of subtasks to generate.
- * @param {boolean} [args.research] - Enable Perplexity AI for research-backed subtask generation.
+ * @param {boolean} [args.research] - Enable research role for subtask generation.
  * @param {string} [args.prompt] - Additional context to guide subtask generation.
  * @param {boolean} [args.force] - Force expansion even if subtasks exist.
+ * @param {string} [args.projectRoot] - Project root directory.
  * @param {Object} log - Logger object
- * @param {Object} context - Context object containing session and reportProgress
+ * @param {Object} context - Context object containing session
+ * @param {Object} [context.session] - MCP Session object
  * @returns {Promise<Object>} - Task expansion result { success: boolean, data?: any, error?: { code: string, message: string }, fromCache: boolean }
  */
 export async function expandTaskDirect(args, log, context = {}) {
-	const { session } = context;
-	// Destructure expected args
-	const { tasksJsonPath, id, num, research, prompt, force } = args;
+	const { session } = context; // Extract session
+	// Destructure expected args, including projectRoot
+	const { tasksJsonPath, id, num, research, prompt, force, projectRoot } = args;
 
 	// Log session root data for debugging
 	log.info(
@@ -85,28 +84,9 @@ export async function expandTaskDirect(args, log, context = {}) {
 	const additionalContext = prompt || '';
 	const forceFlag = force === true;
 
-	// Initialize AI client if needed (for expandTask function)
-	try {
-		// This ensures the AI client is available by checking it
-		if (useResearch) {
-			log.info('Verifying AI client for research-backed expansion');
-			await getAnthropicClientForMCP(session, log);
-		}
-	} catch (error) {
-		log.error(`Failed to initialize AI client: ${error.message}`);
-		return {
-			success: false,
-			error: {
-				code: 'AI_CLIENT_ERROR',
-				message: `Cannot initialize AI client: ${error.message}`
-			},
-			fromCache: false
-		};
-	}
-
 	try {
 		log.info(
-			`[expandTaskDirect] Expanding task ${taskId} into ${numSubtasks || 'default'} subtasks. Research: ${useResearch}`
+			`[expandTaskDirect] Expanding task ${taskId} into ${numSubtasks || 'default'} subtasks. Research: ${useResearch}, Force: ${forceFlag}`
 		);
 
 		// Read tasks data
@@ -202,23 +182,29 @@ export async function expandTaskDirect(args, log, context = {}) {
 		// Save tasks.json with potentially empty subtasks array
 		writeJSON(tasksPath, data);
 
+		// Create logger wrapper using the utility
+		const mcpLog = createLogWrapper(log);
+
+		let wasSilent; // Declare wasSilent outside the try block
 		// Process the request
 		try {
 			// Enable silent mode to prevent console logs from interfering with JSON response
-			enableSilentMode();
+			wasSilent = isSilentMode(); // Assign inside the try block
+			if (!wasSilent) enableSilentMode();
 
-			// Call expandTask with session context to ensure AI client is properly initialized
-			const result = await expandTask(
+			// Call the core expandTask function with the wrapped logger and projectRoot
+			const updatedTaskResult = await expandTask(
 				tasksPath,
 				taskId,
 				numSubtasks,
 				useResearch,
 				additionalContext,
-				{ mcpLog: log, session } // Only pass mcpLog and session, NOT reportProgress
+				{ mcpLog, session, projectRoot },
+				forceFlag
 			);
 
 			// Restore normal logging
-			disableSilentMode();
+			if (!wasSilent && isSilentMode()) disableSilentMode();
 
 			// Read the updated data
 			const updatedData = readJSON(tasksPath);
@@ -244,7 +230,7 @@ export async function expandTaskDirect(args, log, context = {}) {
 			};
 		} catch (error) {
 			// Make sure to restore normal logging even if there's an error
-			disableSilentMode();
+			if (!wasSilent && isSilentMode()) disableSilentMode();
 
 			log.error(`Error expanding task: ${error.message}`);
 			return {

mcp-server/src/core/direct-functions/generate-task-files.js

@@ -8,7 +8,6 @@ import {
 	enableSilentMode,
 	disableSilentMode
 } from '../../../../scripts/modules/utils.js';
-import path from 'path';
 
 /**
  * Direct function wrapper for generateTaskFiles with error handling.

mcp-server/src/core/direct-functions/initialize-project.js

@@ -4,72 +4,43 @@ import {
 	disableSilentMode
 	// isSilentMode // Not used directly here
 } from '../../../../scripts/modules/utils.js';
-import { getProjectRootFromSession } from '../../tools/utils.js'; // Adjust path if necessary
 import os from 'os'; // Import os module for home directory check
 
 /**
  * Direct function wrapper for initializing a project.
  * Derives target directory from session, sets CWD, and calls core init logic.
- * @param {object} args - Arguments containing project details and options (projectName, projectDescription, yes, etc.)
+ * @param {object} args - Arguments containing initialization options (addAliases, skipInstall, yes, projectRoot)
  * @param {object} log - The FastMCP logger instance.
  * @param {object} context - The context object, must contain { session }.
  * @returns {Promise<{success: boolean, data?: any, error?: {code: string, message: string}}>} - Standard result object.
  */
 export async function initializeProjectDirect(args, log, context = {}) {
-	const { session } = context;
+	const { session } = context; // Keep session if core logic needs it
 	const homeDir = os.homedir();
-	let targetDirectory = null;
 
-	log.info(
-		`CONTEXT received in direct function: ${context ? JSON.stringify(Object.keys(context)) : 'MISSING or Falsy'}`
-	);
-	log.info(
-		`SESSION extracted in direct function: ${session ? 'Exists' : 'MISSING or Falsy'}`
-	);
 	log.info(`Args received in direct function: ${JSON.stringify(args)}`);
 
 	// --- Determine Target Directory ---
-	// 1. Prioritize projectRoot passed directly in args
-	// Ensure it's not null, '/', or the home directory
-	if (
-		args.projectRoot &&
-		args.projectRoot !== '/' &&
-		args.projectRoot !== homeDir
-	) {
-		log.info(`Using projectRoot directly from args: ${args.projectRoot}`);
-		targetDirectory = args.projectRoot;
-	} else {
-		// 2. If args.projectRoot is missing or invalid, THEN try session (as a fallback)
-		log.warn(
-			`args.projectRoot ('${args.projectRoot}') is missing or invalid. Attempting to derive from session.`
-		);
-		const sessionDerivedPath = getProjectRootFromSession(session, log);
-		// Validate the session-derived path as well
-		if (
-			sessionDerivedPath &&
-			sessionDerivedPath !== '/' &&
-			sessionDerivedPath !== homeDir
-		) {
-			log.info(
-				`Using project root derived from session: ${sessionDerivedPath}`
-			);
-			targetDirectory = sessionDerivedPath;
-		} else {
-			log.error(
-				`Could not determine a valid project root. args.projectRoot='${args.projectRoot}', sessionDerivedPath='${sessionDerivedPath}'`
-			);
-		}
-	}
+	// TRUST the projectRoot passed from the tool layer via args
+	// The HOF in the tool layer already normalized and validated it came from a reliable source (args or session)
+	const targetDirectory = args.projectRoot;
 
-	// 3. Validate the final targetDirectory
-	if (!targetDirectory) {
-		// This error now covers cases where neither args.projectRoot nor session provided a valid path
+	// --- Validate the targetDirectory (basic sanity checks) ---
+	if (
+		!targetDirectory ||
+		typeof targetDirectory !== 'string' || // Ensure it's a string
+		targetDirectory === '/' ||
+		targetDirectory === homeDir
+	) {
+		log.error(
+			`Invalid target directory received from tool layer: '${targetDirectory}'`
+		);
 		return {
 			success: false,
 			error: {
 				code: 'INVALID_TARGET_DIRECTORY',
-				message: `Cannot initialize project: Could not determine a valid target directory. Please ensure a workspace/folder is open or specify projectRoot.`,
-				details: `Attempted args.projectRoot: ${args.projectRoot}`
+				message: `Cannot initialize project: Invalid target directory '${targetDirectory}' received. Please ensure a valid workspace/folder is open or specified.`,
+				details: `Received args.projectRoot: ${args.projectRoot}` // Show what was received
 			},
 			fromCache: false
 		};
@@ -86,30 +57,26 @@ export async function initializeProjectDirect(args, log, context = {}) {
 	log.info(
 		`Temporarily changing CWD to ${targetDirectory} for initialization.`
 	);
-	process.chdir(targetDirectory); // Change CWD to the *validated* targetDirectory
+	process.chdir(targetDirectory); // Change CWD to the HOF-provided root
 
-	enableSilentMode(); // Enable silent mode BEFORE calling the core function
+	enableSilentMode();
 	try {
-		// Always force yes: true when called via MCP to avoid interactive prompts
+		// Construct options ONLY from the relevant flags in args
+		// The core initializeProject operates in the current CWD, which we just set
 		const options = {
-			name: args.projectName,
-			description: args.projectDescription,
-			version: args.projectVersion,
-			author: args.authorName,
-			skipInstall: args.skipInstall,
 			aliases: args.addAliases,
+			skipInstall: args.skipInstall,
 			yes: true // Force yes mode
 		};
 
 		log.info(`Initializing project with options: ${JSON.stringify(options)}`);
 		const result = await initializeProject(options); // Call core logic
 
-		// Format success result for handleApiResult
 		resultData = {
 			message: 'Project initialized successfully.',
 			next_step:
 				'Now that the project is initialized, the next step is to create the tasks by parsing a PRD. This will create the tasks folder and the initial task files (tasks folder will be created when parse-prd is run). The parse-prd tool will require a prd.txt file as input (typically found in the project root directory, scripts/ directory). You can create a prd.txt file by asking the user about their idea, and then using the scripts/example_prd.txt file as a template to genrate a prd.txt file in scripts/. You may skip all of this if the user already has a prd.txt file. You can THEN use the parse-prd tool to create the tasks. So: step 1 after initialization is to create a prd.txt file in scripts/prd.txt or confirm the user already has one. Step 2 is to use the parse-prd tool to create the tasks. Do not bother looking for tasks after initialization, just use the parse-prd tool to create the tasks after creating a prd.txt from which to parse the tasks. You do NOT need to reinitialize the project to parse-prd.',
-			...result // Include details returned by initializeProject
+			...result
 		};
 		success = true;
 		log.info(
@@ -124,12 +91,11 @@ export async function initializeProjectDirect(args, log, context = {}) {
 		};
 		success = false;
 	} finally {
-		disableSilentMode(); // ALWAYS disable silent mode in finally
+		disableSilentMode();
 		log.info(`Restoring original CWD: ${originalCwd}`);
-		process.chdir(originalCwd); // Change back to original CWD
+		process.chdir(originalCwd);
 	}
 
-	// Return in format expected by handleApiResult
 	if (success) {
 		return { success: true, data: resultData, fromCache: false };
 	} else {

mcp-server/src/core/direct-functions/models.js

@@ -0,0 +1,121 @@
+/**
+ * models.js
+ * Direct function for managing AI model configurations via MCP
+ */
+
+import {
+	getModelConfiguration,
+	getAvailableModelsList,
+	setModel
+} from '../../../../scripts/modules/task-manager/models.js';
+import {
+	enableSilentMode,
+	disableSilentMode
+} from '../../../../scripts/modules/utils.js';
+import { createLogWrapper } from '../../tools/utils.js';
+
+/**
+ * Get or update model configuration
+ * @param {Object} args - Arguments passed by the MCP tool
+ * @param {Object} log - MCP logger
+ * @param {Object} context - MCP context (contains session)
+ * @returns {Object} Result object with success, data/error fields
+ */
+export async function modelsDirect(args, log, context = {}) {
+	const { session } = context;
+	const { projectRoot } = args; // Extract projectRoot from args
+
+	// Create a logger wrapper that the core functions can use
+	const mcpLog = createLogWrapper(log);
+
+	log.info(`Executing models_direct with args: ${JSON.stringify(args)}`);
+	log.info(`Using project root: ${projectRoot}`);
+
+	// Validate flags: cannot use both openrouter and ollama simultaneously
+	if (args.openrouter && args.ollama) {
+		log.error(
+			'Error: Cannot use both openrouter and ollama flags simultaneously.'
+		);
+		return {
+			success: false,
+			error: {
+				code: 'INVALID_ARGS',
+				message: 'Cannot use both openrouter and ollama flags simultaneously.'
+			}
+		};
+	}
+
+	try {
+		enableSilentMode();
+
+		try {
+			// Check for the listAvailableModels flag
+			if (args.listAvailableModels === true) {
+				return await getAvailableModelsList({
+					session,
+					mcpLog,
+					projectRoot // Pass projectRoot to function
+				});
+			}
+
+			// Handle setting a specific model
+			if (args.setMain) {
+				return await setModel('main', args.setMain, {
+					session,
+					mcpLog,
+					projectRoot, // Pass projectRoot to function
+					providerHint: args.openrouter
+						? 'openrouter'
+						: args.ollama
+							? 'ollama'
+							: undefined // Pass hint
+				});
+			}
+
+			if (args.setResearch) {
+				return await setModel('research', args.setResearch, {
+					session,
+					mcpLog,
+					projectRoot, // Pass projectRoot to function
+					providerHint: args.openrouter
+						? 'openrouter'
+						: args.ollama
+							? 'ollama'
+							: undefined // Pass hint
+				});
+			}
+
+			if (args.setFallback) {
+				return await setModel('fallback', args.setFallback, {
+					session,
+					mcpLog,
+					projectRoot, // Pass projectRoot to function
+					providerHint: args.openrouter
+						? 'openrouter'
+						: args.ollama
+							? 'ollama'
+							: undefined // Pass hint
+				});
+			}
+
+			// Default action: get current configuration
+			return await getModelConfiguration({
+				session,
+				mcpLog,
+				projectRoot // Pass projectRoot to function
+			});
+		} finally {
+			disableSilentMode();
+		}
+	} catch (error) {
+		log.error(`Error in models_direct: ${error.message}`);
+		return {
+			success: false,
+			error: {
+				code: 'DIRECT_FUNCTION_ERROR',
+				message: error.message,
+				details: error.stack
+			}
+		};
+	}
+}

mcp-server/src/core/direct-functions/next-task.js

@@ -71,24 +71,34 @@ export async function nextTaskDirect(args, log) {
 					data: {
 						message:
 							'No eligible next task found. All tasks are either completed or have unsatisfied dependencies',
-						nextTask: null,
-						allTasks: data.tasks
+						nextTask: null
 					}
 				};
 			}
 
+			// Check if it's a subtask
+			const isSubtask =
+				typeof nextTask.id === 'string' && nextTask.id.includes('.');
+
+			const taskOrSubtask = isSubtask ? 'subtask' : 'task';
+
+			const additionalAdvice = isSubtask
+				? 'Subtasks can be updated with timestamped details as you implement them. This is useful for tracking progress, marking milestones and insights (of successful or successive falures in attempting to implement the subtask). Research can be used when updating the subtask to collect up-to-date information, and can be helpful to solve a repeating problem the agent is unable to solve. It is a good idea to get-task the parent task to collect the overall context of the task, and to get-task the subtask to collect the specific details of the subtask.'
+				: 'Tasks can be updated to reflect a change in the direction of the task, or to reformulate the task per your prompt. Research can be used when updating the task to collect up-to-date information. It is best to update subtasks as you work on them, and to update the task for more high-level changes that may affect pending subtasks or the general direction of the task.';
+
 			// Restore normal logging
 			disableSilentMode();
 
 			// Return the next task data with the full tasks array for reference
 			log.info(
-				`Successfully found next task ${nextTask.id}: ${nextTask.title}`
+				`Successfully found next task ${nextTask.id}: ${nextTask.title}. Is subtask: ${isSubtask}`
 			);
 			return {
 				success: true,
 				data: {
 					nextTask,
-					allTasks: data.tasks
+					isSubtask,
+					nextSteps: `When ready to work on the ${taskOrSubtask}, use set-status to set the status to "in progress" ${additionalAdvice}`
 				}
 			};
 		} catch (error) {

mcp-server/src/core/direct-functions/parse-prd.js

@@ -5,206 +5,176 @@
 
 import path from 'path';
 import fs from 'fs';
-import os from 'os'; // Import os module for home directory check
 import { parsePRD } from '../../../../scripts/modules/task-manager.js';
-import { findTasksJsonPath } from '../utils/path-utils.js';
 import {
 	enableSilentMode,
-	disableSilentMode
+	disableSilentMode,
+	isSilentMode
 } from '../../../../scripts/modules/utils.js';
-import {
-	getAnthropicClientForMCP,
-	getModelConfig
-} from '../utils/ai-client-utils.js';
+import { createLogWrapper } from '../../tools/utils.js';
+import { getDefaultNumTasks } from '../../../../scripts/modules/config-manager.js';
 
 /**
  * Direct function wrapper for parsing PRD documents and generating tasks.
  *
- * @param {Object} args - Command arguments containing input, numTasks or tasks, and output options.
+ * @param {Object} args - Command arguments containing projectRoot, input, output, numTasks options.
  * @param {Object} log - Logger object.
  * @param {Object} context - Context object containing session data.
  * @returns {Promise<Object>} - Result object with success status and data/error information.
  */
 export async function parsePRDDirect(args, log, context = {}) {
-	const { session } = context; // Only extract session, not reportProgress
+	const { session } = context;
+	// Extract projectRoot from args
+	const {
+		input: inputArg,
+		output: outputArg,
+		numTasks: numTasksArg,
+		force,
+		append,
+		projectRoot
+	} = args;
 
-	try {
-		log.info(`Parsing PRD document with args: ${JSON.stringify(args)}`);
+	// Create the standard logger wrapper
+	const logWrapper = createLogWrapper(log);
 
-		// Initialize AI client for PRD parsing
-		let aiClient;
-		try {
-			aiClient = getAnthropicClientForMCP(session, log);
-		} catch (error) {
-			log.error(`Failed to initialize AI client: ${error.message}`);
-			return {
-				success: false,
-				error: {
-					code: 'AI_CLIENT_ERROR',
-					message: `Cannot initialize AI client: ${error.message}`
-				},
-				fromCache: false
-			};
-		}
-
-		// Validate required parameters
-		if (!args.projectRoot) {
-			const errorMessage = 'Project root is required for parsePRDDirect';
-			log.error(errorMessage);
-			return {
-				success: false,
-				error: { code: 'MISSING_PROJECT_ROOT', message: errorMessage },
-				fromCache: false
-			};
-		}
-
-		if (!args.input) {
-			const errorMessage = 'Input file path is required for parsePRDDirect';
-			log.error(errorMessage);
-			return {
-				success: false,
-				error: { code: 'MISSING_INPUT_PATH', message: errorMessage },
-				fromCache: false
-			};
-		}
-
-		if (!args.output) {
-			const errorMessage = 'Output file path is required for parsePRDDirect';
-			log.error(errorMessage);
-			return {
-				success: false,
-				error: { code: 'MISSING_OUTPUT_PATH', message: errorMessage },
-				fromCache: false
-			};
-		}
-
-		// Resolve input path (expecting absolute path or path relative to project root)
-		const projectRoot = args.projectRoot;
-		const inputPath = path.isAbsolute(args.input)
-			? args.input
-			: path.resolve(projectRoot, args.input);
-
-		// Verify input file exists
-		if (!fs.existsSync(inputPath)) {
-			const errorMessage = `Input file not found: ${inputPath}`;
-			log.error(errorMessage);
-			return {
-				success: false,
-				error: {
-					code: 'INPUT_FILE_NOT_FOUND',
-					message: errorMessage,
-					details: `Checked path: ${inputPath}\nProject root: ${projectRoot}\nInput argument: ${args.input}`
-				},
-				fromCache: false
-			};
-		}
-
-		// Resolve output path (expecting absolute path or path relative to project root)
-		const outputPath = path.isAbsolute(args.output)
-			? args.output
-			: path.resolve(projectRoot, args.output);
-
-		// Ensure output directory exists
-		const outputDir = path.dirname(outputPath);
-		if (!fs.existsSync(outputDir)) {
-			log.info(`Creating output directory: ${outputDir}`);
-			fs.mkdirSync(outputDir, { recursive: true });
-		}
-
-		// Parse number of tasks - handle both string and number values
-		let numTasks = 10; // Default
-		if (args.numTasks) {
-			numTasks =
-				typeof args.numTasks === 'string'
-					? parseInt(args.numTasks, 10)
-					: args.numTasks;
-			if (isNaN(numTasks)) {
-				numTasks = 10; // Fallback to default if parsing fails
-				log.warn(`Invalid numTasks value: ${args.numTasks}. Using default: 10`);
-			}
-		}
-
-		log.info(
-			`Preparing to parse PRD from ${inputPath} and output to ${outputPath} with ${numTasks} tasks`
-		);
-
-		// Create the logger wrapper for proper logging in the core function
-		const logWrapper = {
-			info: (message, ...args) => log.info(message, ...args),
-			warn: (message, ...args) => log.warn(message, ...args),
-			error: (message, ...args) => log.error(message, ...args),
-			debug: (message, ...args) => log.debug && log.debug(message, ...args),
-			success: (message, ...args) => log.info(message, ...args) // Map success to info
-		};
-
-		// Get model config from session
-		const modelConfig = getModelConfig(session);
-
-		// Enable silent mode to prevent console logs from interfering with JSON response
-		enableSilentMode();
-		try {
-			// Make sure the output directory exists
-			const outputDir = path.dirname(outputPath);
-			if (!fs.existsSync(outputDir)) {
-				log.info(`Creating output directory: ${outputDir}`);
-				fs.mkdirSync(outputDir, { recursive: true });
-			}
-
-			// Execute core parsePRD function with AI client
-			await parsePRD(
-				inputPath,
-				outputPath,
-				numTasks,
-				{
-					mcpLog: logWrapper,
-					session
-				},
-				aiClient,
-				modelConfig
-			);
-
-			// Since parsePRD doesn't return a value but writes to a file, we'll read the result
-			// to return it to the caller
-			if (fs.existsSync(outputPath)) {
-				const tasksData = JSON.parse(fs.readFileSync(outputPath, 'utf8'));
-				log.info(
-					`Successfully parsed PRD and generated ${tasksData.tasks?.length || 0} tasks`
-				);
-
-				return {
-					success: true,
-					data: {
-						message: `Successfully generated ${tasksData.tasks?.length || 0} tasks from PRD`,
-						taskCount: tasksData.tasks?.length || 0,
-						outputPath
-					},
-					fromCache: false // This operation always modifies state and should never be cached
-				};
-			} else {
-				const errorMessage = `Tasks file was not created at ${outputPath}`;
-				log.error(errorMessage);
-				return {
-					success: false,
-					error: { code: 'OUTPUT_FILE_NOT_CREATED', message: errorMessage },
-					fromCache: false
-				};
-			}
-		} finally {
-			// Always restore normal logging
-			disableSilentMode();
-		}
-	} catch (error) {
-		// Make sure to restore normal logging even if there's an error
-		disableSilentMode();
-
-		log.error(`Error parsing PRD: ${error.message}`);
+	// --- Input Validation and Path Resolution ---
+	if (!projectRoot) {
+		logWrapper.error('parsePRDDirect requires a projectRoot argument.');
 		return {
 			success: false,
 			error: {
-				code: 'PARSE_PRD_ERROR',
-				message: error.message || 'Unknown error parsing PRD'
-			},
-			fromCache: false
+				code: 'MISSING_ARGUMENT',
+				message: 'projectRoot is required.'
+			}
 		};
 	}
+	if (!inputArg) {
+		logWrapper.error('parsePRDDirect called without input path');
+		return {
+			success: false,
+			error: { code: 'MISSING_ARGUMENT', message: 'Input path is required' }
+		};
+	}
+
+	// Resolve input and output paths relative to projectRoot
+	const inputPath = path.resolve(projectRoot, inputArg);
+	const outputPath = outputArg
+		? path.resolve(projectRoot, outputArg)
+		: path.resolve(projectRoot, 'tasks', 'tasks.json'); // Default output path
+
+	// Check if input file exists
+	if (!fs.existsSync(inputPath)) {
+		const errorMsg = `Input PRD file not found at resolved path: ${inputPath}`;
+		logWrapper.error(errorMsg);
+		return {
+			success: false,
+			error: { code: 'FILE_NOT_FOUND', message: errorMsg }
+		};
+	}
+
+	const outputDir = path.dirname(outputPath);
+	try {
+		if (!fs.existsSync(outputDir)) {
+			logWrapper.info(`Creating output directory: ${outputDir}`);
+			fs.mkdirSync(outputDir, { recursive: true });
+		}
+	} catch (dirError) {
+		logWrapper.error(
+			`Failed to create output directory ${outputDir}: ${dirError.message}`
+		);
+		// Return an error response immediately if dir creation fails
+		return {
+			success: false,
+			error: {
+				code: 'DIRECTORY_CREATION_ERROR',
+				message: `Failed to create output directory: ${dirError.message}`
+			}
+		};
+	}
+
+	let numTasks = getDefaultNumTasks(projectRoot);
+	if (numTasksArg) {
+		numTasks =
+			typeof numTasksArg === 'string' ? parseInt(numTasksArg, 10) : numTasksArg;
+		if (isNaN(numTasks) || numTasks <= 0) {
+			// Ensure positive number
+			numTasks = getDefaultNumTasks(projectRoot); // Fallback to default if parsing fails or invalid
+			logWrapper.warn(
+				`Invalid numTasks value: ${numTasksArg}. Using default: ${numTasks}`
+			);
+		}
+	}
+
+	const useForce = force === true;
+	const useAppend = append === true;
+	if (useAppend) {
+		logWrapper.info('Append mode enabled.');
+		if (useForce) {
+			logWrapper.warn(
+				'Both --force and --append flags were provided. --force takes precedence; append mode will be ignored.'
+			);
+		}
+	}
+
+	logWrapper.info(
+		`Parsing PRD via direct function. Input: ${inputPath}, Output: ${outputPath}, NumTasks: ${numTasks}, Force: ${useForce}, Append: ${useAppend}, ProjectRoot: ${projectRoot}`
+	);
+
+	const wasSilent = isSilentMode();
+	if (!wasSilent) {
+		enableSilentMode();
+	}
+
+	try {
+		// Call the core parsePRD function
+		const result = await parsePRD(
+			inputPath,
+			outputPath,
+			numTasks,
+			{ session, mcpLog: logWrapper, projectRoot, useForce, useAppend },
+			'json'
+		);
+
+		// parsePRD returns { success: true, tasks: processedTasks } on success
+		if (result && result.success && Array.isArray(result.tasks)) {
+			logWrapper.success(
+				`Successfully parsed PRD. Generated ${result.tasks.length} tasks.`
+			);
+			return {
+				success: true,
+				data: {
+					message: `Successfully parsed PRD and generated ${result.tasks.length} tasks.`,
+					outputPath: outputPath,
+					taskCount: result.tasks.length
+				}
+			};
+		} else {
+			// Handle case where core function didn't return expected success structure
+			logWrapper.error(
+				'Core parsePRD function did not return a successful structure.'
+			);
+			return {
+				success: false,
+				error: {
+					code: 'CORE_FUNCTION_ERROR',
+					message:
+						result?.message ||
+						'Core function failed to parse PRD or returned unexpected result.'
+				}
+			};
+		}
+	} catch (error) {
+		logWrapper.error(`Error executing core parsePRD: ${error.message}`);
+		return {
+			success: false,
+			error: {
+				code: 'PARSE_PRD_CORE_ERROR',
+				message: error.message || 'Unknown error parsing PRD'
+			}
+		};
+	} finally {
+		if (!wasSilent && isSilentMode()) {
+			disableSilentMode();
+		}
+	}
 }

mcp-server/src/core/direct-functions/remove-task.js

@@ -3,18 +3,23 @@
  * Direct function implementation for removing a task
  */
 
-import { removeTask } from '../../../../scripts/modules/task-manager.js';
+import {
+	removeTask,
+	taskExists
+} from '../../../../scripts/modules/task-manager.js';
 import {
 	enableSilentMode,
-	disableSilentMode
+	disableSilentMode,
+	readJSON
 } from '../../../../scripts/modules/utils.js';
 
 /**
  * Direct function wrapper for removeTask with error handling.
+ * Supports removing multiple tasks at once with comma-separated IDs.
  *
  * @param {Object} args - Command arguments
  * @param {string} args.tasksJsonPath - Explicit path to the tasks.json file.
- * @param {string} args.id - The ID of the task or subtask to remove.
+ * @param {string} args.id - The ID(s) of the task(s) or subtask(s) to remove (comma-separated for multiple).
  * @param {Object} log - Logger object
  * @returns {Promise<Object>} - Remove task result { success: boolean, data?: any, error?: { code: string, message: string }, fromCache: false }
  */
@@ -36,8 +41,7 @@ export async function removeTaskDirect(args, log) {
 		}
 
 		// Validate task ID parameter
-		const taskId = id;
-		if (!taskId) {
+		if (!id) {
 			log.error('Task ID is required');
 			return {
 				success: false,
@@ -49,46 +53,103 @@ export async function removeTaskDirect(args, log) {
 			};
 		}
 
-		// Skip confirmation in the direct function since it's handled by the client
-		log.info(`Removing task with ID: ${taskId} from ${tasksJsonPath}`);
+		// Split task IDs if comma-separated
+		const taskIdArray = id.split(',').map((taskId) => taskId.trim());
 
-		try {
-			// Enable silent mode to prevent console logs from interfering with JSON response
-			enableSilentMode();
+		log.info(
+			`Removing ${taskIdArray.length} task(s) with ID(s): ${taskIdArray.join(', ')} from ${tasksJsonPath}`
+		);
 
-			// Call the core removeTask function using the provided path
-			const result = await removeTask(tasksJsonPath, taskId);
-
-			// Restore normal logging
-			disableSilentMode();
-
-			log.info(`Successfully removed task: ${taskId}`);
-
-			// Return the result
-			return {
-				success: true,
-				data: {
-					message: result.message,
-					taskId: taskId,
-					tasksPath: tasksJsonPath,
-					removedTask: result.removedTask
-				},
-				fromCache: false
-			};
-		} catch (error) {
-			// Make sure to restore normal logging even if there's an error
-			disableSilentMode();
-
-			log.error(`Error removing task: ${error.message}`);
+		// Validate all task IDs exist before proceeding
+		const data = readJSON(tasksJsonPath);
+		if (!data || !data.tasks) {
 			return {
 				success: false,
 				error: {
-					code: error.code || 'REMOVE_TASK_ERROR',
-					message: error.message || 'Failed to remove task'
+					code: 'INVALID_TASKS_FILE',
+					message: `No valid tasks found in ${tasksJsonPath}`
 				},
 				fromCache: false
 			};
 		}
+
+		const invalidTasks = taskIdArray.filter(
+			(taskId) => !taskExists(data.tasks, taskId)
+		);
+
+		if (invalidTasks.length > 0) {
+			return {
+				success: false,
+				error: {
+					code: 'INVALID_TASK_ID',
+					message: `The following tasks were not found: ${invalidTasks.join(', ')}`
+				},
+				fromCache: false
+			};
+		}
+
+		// Remove tasks one by one
+		const results = [];
+
+		// Enable silent mode to prevent console logs from interfering with JSON response
+		enableSilentMode();
+
+		try {
+			for (const taskId of taskIdArray) {
+				try {
+					const result = await removeTask(tasksJsonPath, taskId);
+					results.push({
+						taskId,
+						success: true,
+						message: result.message,
+						removedTask: result.removedTask
+					});
+					log.info(`Successfully removed task: ${taskId}`);
+				} catch (error) {
+					results.push({
+						taskId,
+						success: false,
+						error: error.message
+					});
+					log.error(`Error removing task ${taskId}: ${error.message}`);
+				}
+			}
+		} finally {
+			// Restore normal logging
+			disableSilentMode();
+		}
+
+		// Check if all tasks were successfully removed
+		const successfulRemovals = results.filter((r) => r.success);
+		const failedRemovals = results.filter((r) => !r.success);
+
+		if (successfulRemovals.length === 0) {
+			// All removals failed
+			return {
+				success: false,
+				error: {
+					code: 'REMOVE_TASK_ERROR',
+					message: 'Failed to remove any tasks',
+					details: failedRemovals
+						.map((r) => `${r.taskId}: ${r.error}`)
+						.join('; ')
+				},
+				fromCache: false
+			};
+		}
+
+		// At least some tasks were removed successfully
+		return {
+			success: true,
+			data: {
+				totalTasks: taskIdArray.length,
+				successful: successfulRemovals.length,
+				failed: failedRemovals.length,
+				results: results,
+				tasksPath: tasksJsonPath
+			},
+			fromCache: false
+		};
 	} catch (error) {
 		// Ensure silent mode is disabled even if an outer error occurs
 		disableSilentMode();

mcp-server/src/core/direct-functions/show-task.js

@@ -3,143 +3,100 @@
  * Direct function implementation for showing task details
  */
 
-import { findTaskById } from '../../../../scripts/modules/utils.js';
-import { readJSON } from '../../../../scripts/modules/utils.js';
+import { findTaskById, readJSON } from '../../../../scripts/modules/utils.js';
 import { getCachedOrExecute } from '../../tools/utils.js';
 import {
 	enableSilentMode,
 	disableSilentMode
 } from '../../../../scripts/modules/utils.js';
+import { findTasksJsonPath } from '../utils/path-utils.js';
 
 /**
- * Direct function wrapper for showing task details with error handling and caching.
+ * Direct function wrapper for getting task details.
  *
- * @param {Object} args - Command arguments
- * @param {string} args.tasksJsonPath - Explicit path to the tasks.json file.
- * @param {string} args.id - The ID of the task or subtask to show.
- * @param {Object} log - Logger object
- * @returns {Promise<Object>} - Task details result { success: boolean, data?: any, error?: { code: string, message: string }, fromCache: boolean }
+ * @param {Object} args - Command arguments.
+ * @param {string} args.id - Task ID to show.
+ * @param {string} [args.file] - Optional path to the tasks file (passed to findTasksJsonPath).
+ * @param {string} [args.status] - Optional status to filter subtasks by.
+ * @param {string} args.projectRoot - Absolute path to the project root directory (already normalized by tool).
+ * @param {Object} log - Logger object.
+ * @param {Object} context - Context object containing session data.
+ * @returns {Promise<Object>} - Result object with success status and data/error information.
  */
 export async function showTaskDirect(args, log) {
-	// Destructure expected args
-	const { tasksJsonPath, id } = args;
+	// Destructure session from context if needed later, otherwise ignore
+	// const { session } = context;
+	// Destructure projectRoot and other args. projectRoot is assumed normalized.
+	const { id, file, status, projectRoot } = args;
 
-	if (!tasksJsonPath) {
-		log.error('showTaskDirect called without tasksJsonPath');
+	log.info(
+		`Showing task direct function. ID: ${id}, File: ${file}, Status Filter: ${status}, ProjectRoot: ${projectRoot}`
+	);
+
+	// --- Path Resolution using the passed (already normalized) projectRoot ---
+	let tasksJsonPath;
+	try {
+		// Use the projectRoot passed directly from args
+		tasksJsonPath = findTasksJsonPath(
+			{ projectRoot: projectRoot, file: file },
+			log
+		);
+		log.info(`Resolved tasks path: ${tasksJsonPath}`);
+	} catch (error) {
+		log.error(`Error finding tasks.json: ${error.message}`);
 		return {
 			success: false,
 			error: {
-				code: 'MISSING_ARGUMENT',
-				message: 'tasksJsonPath is required'
-			},
-			fromCache: false
+				code: 'TASKS_FILE_NOT_FOUND',
+				message: `Failed to find tasks.json: ${error.message}`
+			}
 		};
 	}
+	// --- End Path Resolution ---
 
-	// Validate task ID
-	const taskId = id;
-	if (!taskId) {
-		log.error('Task ID is required');
-		return {
-			success: false,
-			error: {
-				code: 'INPUT_VALIDATION_ERROR',
-				message: 'Task ID is required'
-			},
-			fromCache: false
-		};
-	}
-
-	// Generate cache key using the provided task path and ID
-	const cacheKey = `showTask:${tasksJsonPath}:${taskId}`;
-
-	// Define the action function to be executed on cache miss
-	const coreShowTaskAction = async () => {
-		try {
-			// Enable silent mode to prevent console logs from interfering with JSON response
-			enableSilentMode();
-
-			log.info(
-				`Retrieving task details for ID: ${taskId} from ${tasksJsonPath}`
-			);
-
-			// Read tasks data using the provided path
-			const data = readJSON(tasksJsonPath);
-			if (!data || !data.tasks) {
-				disableSilentMode(); // Disable before returning
-				return {
-					success: false,
-					error: {
-						code: 'INVALID_TASKS_FILE',
-						message: `No valid tasks found in ${tasksJsonPath}`
-					}
-				};
-			}
-
-			// Find the specific task
-			const task = findTaskById(data.tasks, taskId);
-
-			if (!task) {
-				disableSilentMode(); // Disable before returning
-				return {
-					success: false,
-					error: {
-						code: 'TASK_NOT_FOUND',
-						message: `Task with ID ${taskId} not found`
-					}
-				};
-			}
-
-			// Restore normal logging
-			disableSilentMode();
-
-			// Return the task data with the full tasks array for reference
-			// (needed for formatDependenciesWithStatus function in UI)
-			log.info(`Successfully found task ${taskId}`);
+	// --- Rest of the function remains the same, using tasksJsonPath ---
+	try {
+		const tasksData = readJSON(tasksJsonPath);
+		if (!tasksData || !tasksData.tasks) {
 			return {
-				success: true,
-				data: {
-					task,
-					allTasks: data.tasks
-				}
+				success: false,
+				error: { code: 'INVALID_TASKS_DATA', message: 'Invalid tasks data' }
 			};
-		} catch (error) {
-			// Make sure to restore normal logging even if there's an error
-			disableSilentMode();
+		}
 
-			log.error(`Error showing task: ${error.message}`);
+		const { task, originalSubtaskCount } = findTaskById(
+			tasksData.tasks,
+			id,
+			status
+		);
+
+		if (!task) {
 			return {
 				success: false,
 				error: {
-					code: 'CORE_FUNCTION_ERROR',
-					message: error.message || 'Failed to show task details'
+					code: 'TASK_NOT_FOUND',
+					message: `Task or subtask with ID ${id} not found`
 				}
 			};
 		}
-	};
 
-	// Use the caching utility
-	try {
-		const result = await getCachedOrExecute({
-			cacheKey,
-			actionFn: coreShowTaskAction,
-			log
-		});
-		log.info(`showTaskDirect completed. From cache: ${result.fromCache}`);
-		return result; // Returns { success, data/error, fromCache }
+		log.info(`Successfully retrieved task ${id}.`);
+
+		const returnData = { ...task };
+		if (originalSubtaskCount !== null) {
+			returnData._originalSubtaskCount = originalSubtaskCount;
+			returnData._subtaskFilter = status;
+		}
+
+		return { success: true, data: returnData };
 	} catch (error) {
-		// Catch unexpected errors from getCachedOrExecute itself
-		disableSilentMode();
-		log.error(
-			`Unexpected error during getCachedOrExecute for showTask: ${error.message}`
-		);
+		log.error(`Error showing task ${id}: ${error.message}`);
 		return {
 			success: false,
 			error: {
-				code: 'UNEXPECTED_ERROR',
+				code: 'TASK_OPERATION_ERROR',
 				message: error.message
-			},
-			fromCache: false
+			}
 		};
 	}
 }

mcp-server/src/core/direct-functions/update-subtask-by-id.js

@@ -6,32 +6,40 @@
 import { updateSubtaskById } from '../../../../scripts/modules/task-manager.js';
 import {
 	enableSilentMode,
-	disableSilentMode
+	disableSilentMode,
+	isSilentMode
 } from '../../../../scripts/modules/utils.js';
-import {
-	getAnthropicClientForMCP,
-	getPerplexityClientForMCP
-} from '../utils/ai-client-utils.js';
+import { createLogWrapper } from '../../tools/utils.js';
 
 /**
  * Direct function wrapper for updateSubtaskById with error handling.
  *
- * @param {Object} args - Command arguments containing id, prompt, useResearch and tasksJsonPath.
+ * @param {Object} args - Command arguments containing id, prompt, useResearch, tasksJsonPath, and projectRoot.
+ * @param {string} args.tasksJsonPath - Explicit path to the tasks.json file.
+ * @param {string} args.id - Subtask ID in format "parent.sub".
+ * @param {string} args.prompt - Information to append to the subtask.
+ * @param {boolean} [args.research] - Whether to use research role.
+ * @param {string} [args.projectRoot] - Project root path.
  * @param {Object} log - Logger object.
  * @param {Object} context - Context object containing session data.
  * @returns {Promise<Object>} - Result object with success status and data/error information.
  */
 export async function updateSubtaskByIdDirect(args, log, context = {}) {
-	const { session } = context; // Only extract session, not reportProgress
-	const { tasksJsonPath, id, prompt, research } = args;
+	const { session } = context;
+	// Destructure expected args, including projectRoot
+	const { tasksJsonPath, id, prompt, research, projectRoot } = args;
+
+	const logWrapper = createLogWrapper(log);
 
 	try {
-		log.info(`Updating subtask with args: ${JSON.stringify(args)}`);
+		logWrapper.info(
+			`Updating subtask by ID via direct function. ID: ${id}, ProjectRoot: ${projectRoot}`
+		);
 
 		// Check if tasksJsonPath was provided
 		if (!tasksJsonPath) {
 			const errorMessage = 'tasksJsonPath is required but was not provided.';
-			log.error(errorMessage);
+			logWrapper.error(errorMessage);
 			return {
 				success: false,
 				error: { code: 'MISSING_ARGUMENT', message: errorMessage },
@@ -39,22 +47,22 @@ export async function updateSubtaskByIdDirect(args, log, context = {}) {
 			};
 		}
 
-		// Check required parameters (id and prompt)
-		if (!id) {
+		// Basic validation for ID format (e.g., '5.2')
+		if (!id || typeof id !== 'string' || !id.includes('.')) {
 			const errorMessage =
-				'No subtask ID specified. Please provide a subtask ID to update.';
-			log.error(errorMessage);
+				'Invalid subtask ID format. Must be in format "parentId.subtaskId" (e.g., "5.2").';
+			logWrapper.error(errorMessage);
 			return {
 				success: false,
-				error: { code: 'MISSING_SUBTASK_ID', message: errorMessage },
+				error: { code: 'INVALID_SUBTASK_ID', message: errorMessage },
 				fromCache: false
 			};
 		}
 
 		if (!prompt) {
 			const errorMessage =
-				'No prompt specified. Please provide a prompt with information to add to the subtask.';
-			log.error(errorMessage);
+				'No prompt specified. Please provide the information to append.';
+			logWrapper.error(errorMessage);
 			return {
 				success: false,
 				error: { code: 'MISSING_PROMPT', message: errorMessage },
@@ -87,79 +95,41 @@ export async function updateSubtaskByIdDirect(args, log, context = {}) {
 
 		// Use the provided path
 		const tasksPath = tasksJsonPath;
-
-		// Get research flag
 		const useResearch = research === true;
 
 		log.info(
 			`Updating subtask with ID ${subtaskIdStr} with prompt "${prompt}" and research: ${useResearch}`
 		);
 
-		// Initialize the appropriate AI client based on research flag
-		try {
-			if (useResearch) {
-				// Initialize Perplexity client
-				await getPerplexityClientForMCP(session);
-			} else {
-				// Initialize Anthropic client
-				await getAnthropicClientForMCP(session);
-			}
-		} catch (error) {
-			log.error(`AI client initialization error: ${error.message}`);
-			return {
-				success: false,
-				error: {
-					code: 'AI_CLIENT_ERROR',
-					message: error.message || 'Failed to initialize AI client'
-				},
-				fromCache: false
-			};
+		const wasSilent = isSilentMode();
+		if (!wasSilent) {
+			enableSilentMode();
 		}
 
 		try {
-			// Enable silent mode to prevent console logs from interfering with JSON response
-			enableSilentMode();
-
-			// Create a logger wrapper object to handle logging without breaking the mcpLog[level] calls
-			// This ensures outputFormat is set to 'json' while still supporting proper logging
-			const logWrapper = {
-				info: (message) => log.info(message),
-				warn: (message) => log.warn(message),
-				error: (message) => log.error(message),
-				debug: (message) => log.debug && log.debug(message),
-				success: (message) => log.info(message) // Map success to info if needed
-			};
-
 			// Execute core updateSubtaskById function
-			// Pass both session and logWrapper as mcpLog to ensure outputFormat is 'json'
 			const updatedSubtask = await updateSubtaskById(
 				tasksPath,
 				subtaskIdStr,
 				prompt,
 				useResearch,
-				{
-					session,
-					mcpLog: logWrapper
-				}
+				{ mcpLog: logWrapper, session, projectRoot },
+				'json'
 			);
 
-			// Restore normal logging
-			disableSilentMode();
-
-			// Handle the case where the subtask couldn't be updated (e.g., already marked as done)
-			if (!updatedSubtask) {
+			if (updatedSubtask === null) {
+				const message = `Subtask ${id} or its parent task not found.`;
+				logWrapper.error(message); // Log as error since it couldn't be found
 				return {
 					success: false,
-					error: {
-						code: 'SUBTASK_UPDATE_FAILED',
-						message:
-							'Failed to update subtask. It may be marked as completed, or another error occurred.'
-					},
+					error: { code: 'SUBTASK_NOT_FOUND', message: message },
 					fromCache: false
 				};
 			}
 
-			// Return the updated subtask information
+			// Subtask updated successfully
+			const successMessage = `Successfully updated subtask with ID ${subtaskIdStr}`;
+			logWrapper.success(successMessage);
 			return {
 				success: true,
 				data: {
@@ -170,23 +140,33 @@ export async function updateSubtaskByIdDirect(args, log, context = {}) {
 					tasksPath,
 					useResearch
 				},
-				fromCache: false // This operation always modifies state and should never be cached
+				fromCache: false
 			};
 		} catch (error) {
-			// Make sure to restore normal logging even if there's an error
-			disableSilentMode();
-			throw error; // Rethrow to be caught by outer catch block
+			logWrapper.error(`Error updating subtask by ID: ${error.message}`);
+			return {
+				success: false,
+				error: {
+					code: 'UPDATE_SUBTASK_CORE_ERROR',
+					message: error.message || 'Unknown error updating subtask'
+				},
+				fromCache: false
+			};
+		} finally {
+			if (!wasSilent && isSilentMode()) {
+				disableSilentMode();
+			}
 		}
 	} catch (error) {
-		// Ensure silent mode is disabled
-		disableSilentMode();
-
-		log.error(`Error updating subtask by ID: ${error.message}`);
+		logWrapper.error(
+			`Setup error in updateSubtaskByIdDirect: ${error.message}`
+		);
+		if (isSilentMode()) disableSilentMode();
 		return {
 			success: false,
 			error: {
-				code: 'UPDATE_SUBTASK_ERROR',
-				message: error.message || 'Unknown error updating subtask'
+				code: 'DIRECT_FUNCTION_SETUP_ERROR',
+				message: error.message || 'Unknown setup error'
 			},
 			fromCache: false
 		};

mcp-server/src/core/direct-functions/update-task-by-id.js

@@ -6,33 +6,40 @@
 import { updateTaskById } from '../../../../scripts/modules/task-manager.js';
 import {
 	enableSilentMode,
-	disableSilentMode
+	disableSilentMode,
+	isSilentMode
 } from '../../../../scripts/modules/utils.js';
-import {
-	getAnthropicClientForMCP,
-	getPerplexityClientForMCP
-} from '../utils/ai-client-utils.js';
+import { createLogWrapper } from '../../tools/utils.js';
 
 /**
  * Direct function wrapper for updateTaskById with error handling.
  *
- * @param {Object} args - Command arguments containing id, prompt, useResearch and tasksJsonPath.
+ * @param {Object} args - Command arguments containing id, prompt, useResearch, tasksJsonPath, and projectRoot.
+ * @param {string} args.tasksJsonPath - Explicit path to the tasks.json file.
+ * @param {string} args.id - Task ID (or subtask ID like "1.2").
+ * @param {string} args.prompt - New information/context prompt.
+ * @param {boolean} [args.research] - Whether to use research role.
+ * @param {string} [args.projectRoot] - Project root path.
  * @param {Object} log - Logger object.
  * @param {Object} context - Context object containing session data.
  * @returns {Promise<Object>} - Result object with success status and data/error information.
  */
 export async function updateTaskByIdDirect(args, log, context = {}) {
-	const { session } = context; // Only extract session, not reportProgress
-	// Destructure expected args, including the resolved tasksJsonPath
-	const { tasksJsonPath, id, prompt, research } = args;
+	const { session } = context;
+	// Destructure expected args, including projectRoot
+	const { tasksJsonPath, id, prompt, research, projectRoot } = args;
+
+	const logWrapper = createLogWrapper(log);
 
 	try {
-		log.info(`Updating task with args: ${JSON.stringify(args)}`);
+		logWrapper.info(
+			`Updating task by ID via direct function. ID: ${id}, ProjectRoot: ${projectRoot}`
+		);
 
 		// Check if tasksJsonPath was provided
 		if (!tasksJsonPath) {
 			const errorMessage = 'tasksJsonPath is required but was not provided.';
-			log.error(errorMessage);
+			logWrapper.error(errorMessage);
 			return {
 				success: false,
 				error: { code: 'MISSING_ARGUMENT', message: errorMessage },
@@ -44,7 +51,7 @@ export async function updateTaskByIdDirect(args, log, context = {}) {
 		if (!id) {
 			const errorMessage =
 				'No task ID specified. Please provide a task ID to update.';
-			log.error(errorMessage);
+			logWrapper.error(errorMessage);
 			return {
 				success: false,
 				error: { code: 'MISSING_TASK_ID', message: errorMessage },
@@ -55,7 +62,7 @@ export async function updateTaskByIdDirect(args, log, context = {}) {
 		if (!prompt) {
 			const errorMessage =
 				'No prompt specified. Please provide a prompt with new information for the task update.';
-			log.error(errorMessage);
+			logWrapper.error(errorMessage);
 			return {
 				success: false,
 				error: { code: 'MISSING_PROMPT', message: errorMessage },
@@ -74,7 +81,7 @@ export async function updateTaskByIdDirect(args, log, context = {}) {
 				taskId = parseInt(id, 10);
 				if (isNaN(taskId)) {
 					const errorMessage = `Invalid task ID: ${id}. Task ID must be a positive integer or subtask ID (e.g., "5.2").`;
-					log.error(errorMessage);
+					logWrapper.error(errorMessage);
 					return {
 						success: false,
 						error: { code: 'INVALID_TASK_ID', message: errorMessage },
@@ -92,94 +99,80 @@ export async function updateTaskByIdDirect(args, log, context = {}) {
 		// Get research flag
 		const useResearch = research === true;
 
-		// Initialize appropriate AI client based on research flag
-		let aiClient;
-		try {
-			if (useResearch) {
-				log.info('Using Perplexity AI for research-backed task update');
-				aiClient = await getPerplexityClientForMCP(session, log);
-			} else {
-				log.info('Using Claude AI for task update');
-				aiClient = getAnthropicClientForMCP(session, log);
-			}
-		} catch (error) {
-			log.error(`Failed to initialize AI client: ${error.message}`);
-			return {
-				success: false,
-				error: {
-					code: 'AI_CLIENT_ERROR',
-					message: `Cannot initialize AI client: ${error.message}`
-				},
-				fromCache: false
-			};
-		}
-
-		log.info(
+		logWrapper.info(
 			`Updating task with ID ${taskId} with prompt "${prompt}" and research: ${useResearch}`
 		);
 
-		try {
-			// Enable silent mode to prevent console logs from interfering with JSON response
+		const wasSilent = isSilentMode();
+		if (!wasSilent) {
 			enableSilentMode();
+		}
 
-			// Create a logger wrapper that matches what updateTaskById expects
-			const logWrapper = {
-				info: (message) => log.info(message),
-				warn: (message) => log.warn(message),
-				error: (message) => log.error(message),
-				debug: (message) => log.debug && log.debug(message),
-				success: (message) => log.info(message) // Map success to info since many loggers don't have success
-			};
-
+		try {
 			// Execute core updateTaskById function with proper parameters
-			await updateTaskById(
+			const updatedTask = await updateTaskById(
 				tasksPath,
 				taskId,
 				prompt,
 				useResearch,
 				{
-					mcpLog: logWrapper, // Use our wrapper object that has the expected method structure
-					session
+					mcpLog: logWrapper,
+					session,
+					projectRoot
 				},
 				'json'
 			);
 
-			// Since updateTaskById doesn't return a value but modifies the tasks file,
-			// we'll return a success message
+			// Check if the core function indicated the task wasn't updated (e.g., status was 'done')
+			if (updatedTask === null) {
+				// Core function logs the reason, just return success with info
+				const message = `Task ${taskId} was not updated (likely already completed).`;
+				logWrapper.info(message);
+				return {
+					success: true,
+					data: { message: message, taskId: taskId, updated: false },
+					fromCache: false
+				};
+			}
+
+			// Task was updated successfully
+			const successMessage = `Successfully updated task with ID ${taskId} based on the prompt`;
+			logWrapper.success(successMessage);
 			return {
 				success: true,
 				data: {
-					message: `Successfully updated task with ID ${taskId} based on the prompt`,
-					taskId,
-					tasksPath: tasksPath, // Return the used path
-					useResearch
+					message: successMessage,
+					taskId: taskId,
+					tasksPath: tasksPath,
+					useResearch: useResearch,
+					updated: true,
+					updatedTask: updatedTask
 				},
-				fromCache: false // This operation always modifies state and should never be cached
+				fromCache: false
 			};
 		} catch (error) {
-			log.error(`Error updating task by ID: ${error.message}`);
+			logWrapper.error(`Error updating task by ID: ${error.message}`);
 			return {
 				success: false,
 				error: {
-					code: 'UPDATE_TASK_ERROR',
+					code: 'UPDATE_TASK_CORE_ERROR',
 					message: error.message || 'Unknown error updating task'
 				},
 				fromCache: false
 			};
 		} finally {
-			// Make sure to restore normal logging even if there's an error
-			disableSilentMode();
+			if (!wasSilent && isSilentMode()) {
+				disableSilentMode();
+			}
 		}
 	} catch (error) {
-		// Ensure silent mode is disabled
-		disableSilentMode();
-
-		log.error(`Error updating task by ID: ${error.message}`);
+		logWrapper.error(`Setup error in updateTaskByIdDirect: ${error.message}`);
+		if (isSilentMode()) disableSilentMode();
 		return {
 			success: false,
 			error: {
-				code: 'UPDATE_TASK_ERROR',
-				message: error.message || 'Unknown error updating task'
+				code: 'DIRECT_FUNCTION_SETUP_ERROR',
+				message: error.message || 'Unknown setup error'
 			},
 			fromCache: false
 		};

mcp-server/src/core/direct-functions/update-tasks.js

@@ -1,187 +1,124 @@
 /**
  * update-tasks.js
- * Direct function implementation for updating tasks based on new context/prompt
+ * Direct function implementation for updating tasks based on new context
  */
 
+import path from 'path';
 import { updateTasks } from '../../../../scripts/modules/task-manager.js';
-import {
-	enableSilentMode,
-	disableSilentMode
-} from '../../../../scripts/modules/utils.js';
-import {
-	getAnthropicClientForMCP,
-	getPerplexityClientForMCP
-} from '../utils/ai-client-utils.js';
+import { createLogWrapper } from '../../tools/utils.js';
 
 /**
- * Direct function wrapper for updating tasks based on new context/prompt.
+ * Direct function wrapper for updating tasks based on new context.
  *
- * @param {Object} args - Command arguments containing fromId, prompt, useResearch and tasksJsonPath.
+ * @param {Object} args - Command arguments containing projectRoot, from, prompt, research options.
  * @param {Object} log - Logger object.
  * @param {Object} context - Context object containing session data.
  * @returns {Promise<Object>} - Result object with success status and data/error information.
  */
 export async function updateTasksDirect(args, log, context = {}) {
-	const { session } = context; // Only extract session, not reportProgress
-	const { tasksJsonPath, from, prompt, research } = args;
+	const { session } = context;
+	const { from, prompt, research, file: fileArg, projectRoot } = args;
 
-	try {
-		log.info(`Updating tasks with args: ${JSON.stringify(args)}`);
+	// Create the standard logger wrapper
+	const logWrapper = createLogWrapper(log);
 
-		// Check if tasksJsonPath was provided
-		if (!tasksJsonPath) {
-			const errorMessage = 'tasksJsonPath is required but was not provided.';
-			log.error(errorMessage);
-			return {
-				success: false,
-				error: { code: 'MISSING_ARGUMENT', message: errorMessage },
-				fromCache: false
-			};
-		}
-
-		// Check for the common mistake of using 'id' instead of 'from'
-		if (args.id !== undefined && from === undefined) {
-			const errorMessage =
-				"You specified 'id' parameter but 'update' requires 'from' parameter. Use 'from' for this tool or use 'update_task' tool if you want to update a single task.";
-			log.error(errorMessage);
-			return {
-				success: false,
-				error: {
-					code: 'PARAMETER_MISMATCH',
-					message: errorMessage,
-					suggestion:
-						"Use 'from' parameter instead of 'id', or use the 'update_task' tool for single task updates"
-				},
-				fromCache: false
-			};
-		}
-
-		// Check required parameters
-		if (!from) {
-			const errorMessage =
-				'No from ID specified. Please provide a task ID to start updating from.';
-			log.error(errorMessage);
-			return {
-				success: false,
-				error: { code: 'MISSING_FROM_ID', message: errorMessage },
-				fromCache: false
-			};
-		}
-
-		if (!prompt) {
-			const errorMessage =
-				'No prompt specified. Please provide a prompt with new context for task updates.';
-			log.error(errorMessage);
-			return {
-				success: false,
-				error: { code: 'MISSING_PROMPT', message: errorMessage },
-				fromCache: false
-			};
-		}
-
-		// Parse fromId - handle both string and number values
-		let fromId;
-		if (typeof from === 'string') {
-			fromId = parseInt(from, 10);
-			if (isNaN(fromId)) {
-				const errorMessage = `Invalid from ID: ${from}. Task ID must be a positive integer.`;
-				log.error(errorMessage);
-				return {
-					success: false,
-					error: { code: 'INVALID_FROM_ID', message: errorMessage },
-					fromCache: false
-				};
-			}
-		} else {
-			fromId = from;
-		}
-
-		// Get research flag
-		const useResearch = research === true;
-
-		// Initialize appropriate AI client based on research flag
-		let aiClient;
-		try {
-			if (useResearch) {
-				log.info('Using Perplexity AI for research-backed task updates');
-				aiClient = await getPerplexityClientForMCP(session, log);
-			} else {
-				log.info('Using Claude AI for task updates');
-				aiClient = getAnthropicClientForMCP(session, log);
-			}
-		} catch (error) {
-			log.error(`Failed to initialize AI client: ${error.message}`);
-			return {
-				success: false,
-				error: {
-					code: 'AI_CLIENT_ERROR',
-					message: `Cannot initialize AI client: ${error.message}`
-				},
-				fromCache: false
-			};
-		}
-
-		log.info(
-			`Updating tasks from ID ${fromId} with prompt "${prompt}" and research: ${useResearch}`
-		);
-
-		// Create the logger wrapper to ensure compatibility with core functions
-		const logWrapper = {
-			info: (message, ...args) => log.info(message, ...args),
-			warn: (message, ...args) => log.warn(message, ...args),
-			error: (message, ...args) => log.error(message, ...args),
-			debug: (message, ...args) => log.debug && log.debug(message, ...args), // Handle optional debug
-			success: (message, ...args) => log.info(message, ...args) // Map success to info if needed
-		};
-
-		try {
-			// Enable silent mode to prevent console logs from interfering with JSON response
-			enableSilentMode();
-
-			// Execute core updateTasks function, passing the AI client and session
-			await updateTasks(tasksJsonPath, fromId, prompt, useResearch, {
-				mcpLog: logWrapper, // Pass the wrapper instead of the raw log object
-				session
-			});
-
-			// Since updateTasks doesn't return a value but modifies the tasks file,
-			// we'll return a success message
-			return {
-				success: true,
-				data: {
-					message: `Successfully updated tasks from ID ${fromId} based on the prompt`,
-					fromId,
-					tasksPath: tasksJsonPath,
-					useResearch
-				},
-				fromCache: false // This operation always modifies state and should never be cached
-			};
-		} catch (error) {
-			log.error(`Error updating tasks: ${error.message}`);
-			return {
-				success: false,
-				error: {
-					code: 'UPDATE_TASKS_ERROR',
-					message: error.message || 'Unknown error updating tasks'
-				},
-				fromCache: false
-			};
-		} finally {
-			// Make sure to restore normal logging even if there's an error
-			disableSilentMode();
-		}
-	} catch (error) {
-		// Ensure silent mode is disabled
-		disableSilentMode();
-
-		log.error(`Error updating tasks: ${error.message}`);
+	// --- Input Validation ---
+	if (!projectRoot) {
+		logWrapper.error('updateTasksDirect requires a projectRoot argument.');
 		return {
 			success: false,
 			error: {
-				code: 'UPDATE_TASKS_ERROR',
-				message: error.message || 'Unknown error updating tasks'
-			},
-			fromCache: false
+				code: 'MISSING_ARGUMENT',
+				message: 'projectRoot is required.'
+			}
 		};
 	}
+
+	if (!from) {
+		logWrapper.error('updateTasksDirect called without from ID');
+		return {
+			success: false,
+			error: {
+				code: 'MISSING_ARGUMENT',
+				message: 'Starting task ID (from) is required'
+			}
+		};
+	}
+
+	if (!prompt) {
+		logWrapper.error('updateTasksDirect called without prompt');
+		return {
+			success: false,
+			error: {
+				code: 'MISSING_ARGUMENT',
+				message: 'Update prompt is required'
+			}
+		};
+	}
+
+	// Resolve tasks file path
+	const tasksFile = fileArg
+		? path.resolve(projectRoot, fileArg)
+		: path.resolve(projectRoot, 'tasks', 'tasks.json');
+
+	logWrapper.info(
+		`Updating tasks via direct function. From: ${from}, Research: ${research}, File: ${tasksFile}, ProjectRoot: ${projectRoot}`
+	);
+
+	enableSilentMode(); // Enable silent mode
+	try {
+		// Call the core updateTasks function
+		const result = await updateTasks(
+			tasksFile,
+			from,
+			prompt,
+			research,
+			{
+				session,
+				mcpLog: logWrapper,
+				projectRoot
+			},
+			'json'
+		);
+
+		// updateTasks returns { success: true, updatedTasks: [...] } on success
+		if (result && result.success && Array.isArray(result.updatedTasks)) {
+			logWrapper.success(
+				`Successfully updated ${result.updatedTasks.length} tasks.`
+			);
+			return {
+				success: true,
+				data: {
+					message: `Successfully updated ${result.updatedTasks.length} tasks.`,
+					tasksFile,
+					updatedCount: result.updatedTasks.length
+				}
+			};
+		} else {
+			// Handle case where core function didn't return expected success structure
+			logWrapper.error(
+				'Core updateTasks function did not return a successful structure.'
+			);
+			return {
+				success: false,
+				error: {
+					code: 'CORE_FUNCTION_ERROR',
+					message:
+						result?.message ||
+						'Core function failed to update tasks or returned unexpected result.'
+				}
+			};
+		}
+	} catch (error) {
+		logWrapper.error(`Error executing core updateTasks: ${error.message}`);
+		return {
+			success: false,
+			error: {
+				code: 'UPDATE_TASKS_CORE_ERROR',
+				message: error.message || 'Unknown error updating tasks'
+			}
+		};
+	} finally {
+		disableSilentMode(); // Ensure silent mode is disabled
+	}
 }

mcp-server/src/core/task-master-core.js

@@ -28,20 +28,12 @@ import { fixDependenciesDirect } from './direct-functions/fix-dependencies.js';
 import { complexityReportDirect } from './direct-functions/complexity-report.js';
 import { addDependencyDirect } from './direct-functions/add-dependency.js';
 import { removeTaskDirect } from './direct-functions/remove-task.js';
-import { initializeProjectDirect } from './direct-functions/initialize-project-direct.js';
+import { initializeProjectDirect } from './direct-functions/initialize-project.js';
+import { modelsDirect } from './direct-functions/models.js';
 
 // Re-export utility functions
 export { findTasksJsonPath } from './utils/path-utils.js';
 
-// Re-export AI client utilities
-export {
-	getAnthropicClientForMCP,
-	getPerplexityClientForMCP,
-	getModelConfig,
-	getBestAvailableAIModel,
-	handleClaudeError
-} from './utils/ai-client-utils.js';
-
 // Use Map for potential future enhancements like introspection or dynamic dispatch
 export const directFunctions = new Map([
 	['listTasksDirect', listTasksDirect],
@@ -66,7 +58,9 @@ export const directFunctions = new Map([
 	['fixDependenciesDirect', fixDependenciesDirect],
 	['complexityReportDirect', complexityReportDirect],
 	['addDependencyDirect', addDependencyDirect],
-	['removeTaskDirect', removeTaskDirect]
+	['removeTaskDirect', removeTaskDirect],
+	['initializeProjectDirect', initializeProjectDirect],
+	['modelsDirect', modelsDirect]
 ]);
 
 // Re-export all direct function implementations
@@ -94,5 +88,6 @@ export {
 	complexityReportDirect,
 	addDependencyDirect,
 	removeTaskDirect,
-	initializeProjectDirect
+	initializeProjectDirect,
+	modelsDirect
 };

mcp-server/src/core/utils/ai-client-utils.js

@@ -1,213 +0,0 @@
-/**
- * ai-client-utils.js
- * Utility functions for initializing AI clients in MCP context
- */
-
-import { Anthropic } from '@anthropic-ai/sdk';
-import dotenv from 'dotenv';
-
-// Load environment variables for CLI mode
-dotenv.config();
-
-// Default model configuration from CLI environment
-const DEFAULT_MODEL_CONFIG = {
-	model: 'claude-3-7-sonnet-20250219',
-	maxTokens: 64000,
-	temperature: 0.2
-};
-
-/**
- * Get an Anthropic client instance initialized with MCP session environment variables
- * @param {Object} [session] - Session object from MCP containing environment variables
- * @param {Object} [log] - Logger object to use (defaults to console)
- * @returns {Anthropic} Anthropic client instance
- * @throws {Error} If API key is missing
- */
-export function getAnthropicClientForMCP(session, log = console) {
-	try {
-		// Extract API key from session.env or fall back to environment variables
-		const apiKey =
-			session?.env?.ANTHROPIC_API_KEY || process.env.ANTHROPIC_API_KEY;
-
-		if (!apiKey) {
-			throw new Error(
-				'ANTHROPIC_API_KEY not found in session environment or process.env'
-			);
-		}
-
-		// Initialize and return a new Anthropic client
-		return new Anthropic({
-			apiKey,
-			defaultHeaders: {
-				'anthropic-beta': 'output-128k-2025-02-19' // Include header for increased token limit
-			}
-		});
-	} catch (error) {
-		log.error(`Failed to initialize Anthropic client: ${error.message}`);
-		throw error;
-	}
-}
-
-/**
- * Get a Perplexity client instance initialized with MCP session environment variables
- * @param {Object} [session] - Session object from MCP containing environment variables
- * @param {Object} [log] - Logger object to use (defaults to console)
- * @returns {OpenAI} OpenAI client configured for Perplexity API
- * @throws {Error} If API key is missing or OpenAI package can't be imported
- */
-export async function getPerplexityClientForMCP(session, log = console) {
-	try {
-		// Extract API key from session.env or fall back to environment variables
-		const apiKey =
-			session?.env?.PERPLEXITY_API_KEY || process.env.PERPLEXITY_API_KEY;
-
-		if (!apiKey) {
-			throw new Error(
-				'PERPLEXITY_API_KEY not found in session environment or process.env'
-			);
-		}
-
-		// Dynamically import OpenAI (it may not be used in all contexts)
-		const { default: OpenAI } = await import('openai');
-
-		// Initialize and return a new OpenAI client configured for Perplexity
-		return new OpenAI({
-			apiKey,
-			baseURL: 'https://api.perplexity.ai'
-		});
-	} catch (error) {
-		log.error(`Failed to initialize Perplexity client: ${error.message}`);
-		throw error;
-	}
-}
-
-/**
- * Get model configuration from session environment or fall back to defaults
- * @param {Object} [session] - Session object from MCP containing environment variables
- * @param {Object} [defaults] - Default model configuration to use if not in session
- * @returns {Object} Model configuration with model, maxTokens, and temperature
- */
-export function getModelConfig(session, defaults = DEFAULT_MODEL_CONFIG) {
-	// Get values from session or fall back to defaults
-	return {
-		model: session?.env?.MODEL || defaults.model,
-		maxTokens: parseInt(session?.env?.MAX_TOKENS || defaults.maxTokens),
-		temperature: parseFloat(session?.env?.TEMPERATURE || defaults.temperature)
-	};
-}
-
-/**
- * Returns the best available AI model based on specified options
- * @param {Object} session - Session object from MCP containing environment variables
- * @param {Object} options - Options for model selection
- * @param {boolean} [options.requiresResearch=false] - Whether the operation requires research capabilities
- * @param {boolean} [options.claudeOverloaded=false] - Whether Claude is currently overloaded
- * @param {Object} [log] - Logger object to use (defaults to console)
- * @returns {Promise<Object>} Selected model info with type and client
- * @throws {Error} If no AI models are available
- */
-export async function getBestAvailableAIModel(
-	session,
-	options = {},
-	log = console
-) {
-	const { requiresResearch = false, claudeOverloaded = false } = options;
-
-	// Test case: When research is needed but no Perplexity, use Claude
-	if (
-		requiresResearch &&
-		!(session?.env?.PERPLEXITY_API_KEY || process.env.PERPLEXITY_API_KEY) &&
-		(session?.env?.ANTHROPIC_API_KEY || process.env.ANTHROPIC_API_KEY)
-	) {
-		try {
-			log.warn('Perplexity not available for research, using Claude');
-			const client = getAnthropicClientForMCP(session, log);
-			return { type: 'claude', client };
-		} catch (error) {
-			log.error(`Claude not available: ${error.message}`);
-			throw new Error('No AI models available for research');
-		}
-	}
-
-	// Regular path: Perplexity for research when available
-	if (
-		requiresResearch &&
-		(session?.env?.PERPLEXITY_API_KEY || process.env.PERPLEXITY_API_KEY)
-	) {
-		try {
-			const client = await getPerplexityClientForMCP(session, log);
-			return { type: 'perplexity', client };
-		} catch (error) {
-			log.warn(`Perplexity not available: ${error.message}`);
-			// Fall through to Claude as backup
-		}
-	}
-
-	// Test case: Claude for overloaded scenario
-	if (
-		claudeOverloaded &&
-		(session?.env?.ANTHROPIC_API_KEY || process.env.ANTHROPIC_API_KEY)
-	) {
-		try {
-			log.warn(
-				'Claude is overloaded but no alternatives are available. Proceeding with Claude anyway.'
-			);
-			const client = getAnthropicClientForMCP(session, log);
-			return { type: 'claude', client };
-		} catch (error) {
-			log.error(
-				`Claude not available despite being overloaded: ${error.message}`
-			);
-			throw new Error('No AI models available');
-		}
-	}
-
-	// Default case: Use Claude when available and not overloaded
-	if (
-		!claudeOverloaded &&
-		(session?.env?.ANTHROPIC_API_KEY || process.env.ANTHROPIC_API_KEY)
-	) {
-		try {
-			const client = getAnthropicClientForMCP(session, log);
-			return { type: 'claude', client };
-		} catch (error) {
-			log.warn(`Claude not available: ${error.message}`);
-			// Fall through to error if no other options
-		}
-	}
-
-	// If we got here, no models were successfully initialized
-	throw new Error('No AI models available. Please check your API keys.');
-}
-
-/**
- * Handle Claude API errors with user-friendly messages
- * @param {Error} error - The error from Claude API
- * @returns {string} User-friendly error message
- */
-export function handleClaudeError(error) {
-	// Check if it's a structured error response
-	if (error.type === 'error' && error.error) {
-		switch (error.error.type) {
-			case 'overloaded_error':
-				return 'Claude is currently experiencing high demand and is overloaded. Please wait a few minutes and try again.';
-			case 'rate_limit_error':
-				return 'You have exceeded the rate limit. Please wait a few minutes before making more requests.';
-			case 'invalid_request_error':
-				return 'There was an issue with the request format. If this persists, please report it as a bug.';
-			default:
-				return `Claude API error: ${error.error.message}`;
-		}
-	}
-
-	// Check for network/timeout errors
-	if (error.message?.toLowerCase().includes('timeout')) {
-		return 'The request to Claude timed out. Please try again.';
-	}
-	if (error.message?.toLowerCase().includes('network')) {
-		return 'There was a network error connecting to Claude. Please check your internet connection and try again.';
-	}
-
-	// Default error message
-	return `Error communicating with Claude: ${error.message}`;
-}

mcp-server/src/core/utils/async-manager.js

@@ -1,251 +0,0 @@
-import { v4 as uuidv4 } from 'uuid';
-
-class AsyncOperationManager {
-	constructor() {
-		this.operations = new Map(); // Stores active operation state
-		this.completedOperations = new Map(); // Stores completed operations
-		this.maxCompletedOperations = 100; // Maximum number of completed operations to store
-		this.listeners = new Map(); // For potential future notifications
-	}
-
-	/**
-	 * Adds an operation to be executed asynchronously.
-	 * @param {Function} operationFn - The async function to execute (e.g., a Direct function).
-	 * @param {Object} args - Arguments to pass to the operationFn.
-	 * @param {Object} context - The MCP tool context { log, reportProgress, session }.
-	 * @returns {string} The unique ID assigned to this operation.
-	 */
-	addOperation(operationFn, args, context) {
-		const operationId = `op-${uuidv4()}`;
-		const operation = {
-			id: operationId,
-			status: 'pending',
-			startTime: Date.now(),
-			endTime: null,
-			result: null,
-			error: null,
-			// Store necessary parts of context, especially log for background execution
-			log: context.log,
-			reportProgress: context.reportProgress, // Pass reportProgress through
-			session: context.session // Pass session through if needed by the operationFn
-		};
-		this.operations.set(operationId, operation);
-		this.log(operationId, 'info', `Operation added.`);
-
-		// Start execution in the background (don't await here)
-		this._runOperation(operationId, operationFn, args, context).catch((err) => {
-			// Catch unexpected errors during the async execution setup itself
-			this.log(
-				operationId,
-				'error',
-				`Critical error starting operation: ${err.message}`,
-				{ stack: err.stack }
-			);
-			operation.status = 'failed';
-			operation.error = {
-				code: 'MANAGER_EXECUTION_ERROR',
-				message: err.message
-			};
-			operation.endTime = Date.now();
-
-			// Move to completed operations
-			this._moveToCompleted(operationId);
-		});
-
-		return operationId;
-	}
-
-	/**
-	 * Internal function to execute the operation.
-	 * @param {string} operationId - The ID of the operation.
-	 * @param {Function} operationFn - The async function to execute.
-	 * @param {Object} args - Arguments for the function.
-	 * @param {Object} context - The original MCP tool context.
-	 */
-	async _runOperation(operationId, operationFn, args, context) {
-		const operation = this.operations.get(operationId);
-		if (!operation) return; // Should not happen
-
-		operation.status = 'running';
-		this.log(operationId, 'info', `Operation running.`);
-		this.emit('statusChanged', { operationId, status: 'running' });
-
-		try {
-			// Pass the necessary context parts to the direct function
-			// The direct function needs to be adapted if it needs reportProgress
-			// We pass the original context's log, plus our wrapped reportProgress
-			const result = await operationFn(args, operation.log, {
-				reportProgress: (progress) =>
-					this._handleProgress(operationId, progress),
-				mcpLog: operation.log, // Pass log as mcpLog if direct fn expects it
-				session: operation.session
-			});
-
-			operation.status = result.success ? 'completed' : 'failed';
-			operation.result = result.success ? result.data : null;
-			operation.error = result.success ? null : result.error;
-			this.log(
-				operationId,
-				'info',
-				`Operation finished with status: ${operation.status}`
-			);
-		} catch (error) {
-			this.log(
-				operationId,
-				'error',
-				`Operation failed with error: ${error.message}`,
-				{ stack: error.stack }
-			);
-			operation.status = 'failed';
-			operation.error = {
-				code: 'OPERATION_EXECUTION_ERROR',
-				message: error.message
-			};
-		} finally {
-			operation.endTime = Date.now();
-			this.emit('statusChanged', {
-				operationId,
-				status: operation.status,
-				result: operation.result,
-				error: operation.error
-			});
-
-			// Move to completed operations if done or failed
-			if (operation.status === 'completed' || operation.status === 'failed') {
-				this._moveToCompleted(operationId);
-			}
-		}
-	}
-
-	/**
-	 * Move an operation from active operations to completed operations history.
-	 * @param {string} operationId - The ID of the operation to move.
-	 * @private
-	 */
-	_moveToCompleted(operationId) {
-		const operation = this.operations.get(operationId);
-		if (!operation) return;
-
-		// Store only the necessary data in completed operations
-		const completedData = {
-			id: operation.id,
-			status: operation.status,
-			startTime: operation.startTime,
-			endTime: operation.endTime,
-			result: operation.result,
-			error: operation.error
-		};
-
-		this.completedOperations.set(operationId, completedData);
-		this.operations.delete(operationId);
-
-		// Trim completed operations if exceeding maximum
-		if (this.completedOperations.size > this.maxCompletedOperations) {
-			// Get the oldest operation (sorted by endTime)
-			const oldest = [...this.completedOperations.entries()].sort(
-				(a, b) => a[1].endTime - b[1].endTime
-			)[0];
-
-			if (oldest) {
-				this.completedOperations.delete(oldest[0]);
-			}
-		}
-	}
-
-	/**
-	 * Handles progress updates from the running operation and forwards them.
-	 * @param {string} operationId - The ID of the operation reporting progress.
-	 * @param {Object} progress - The progress object { progress, total? }.
-	 */
-	_handleProgress(operationId, progress) {
-		const operation = this.operations.get(operationId);
-		if (operation && operation.reportProgress) {
-			try {
-				// Use the reportProgress function captured from the original context
-				operation.reportProgress(progress);
-				this.log(
-					operationId,
-					'debug',
-					`Reported progress: ${JSON.stringify(progress)}`
-				);
-			} catch (err) {
-				this.log(
-					operationId,
-					'warn',
-					`Failed to report progress: ${err.message}`
-				);
-				// Don't stop the operation, just log the reporting failure
-			}
-		}
-	}
-
-	/**
-	 * Retrieves the status and result/error of an operation.
-	 * @param {string} operationId - The ID of the operation.
-	 * @returns {Object | null} The operation details or null if not found.
-	 */
-	getStatus(operationId) {
-		// First check active operations
-		const operation = this.operations.get(operationId);
-		if (operation) {
-			return {
-				id: operation.id,
-				status: operation.status,
-				startTime: operation.startTime,
-				endTime: operation.endTime,
-				result: operation.result,
-				error: operation.error
-			};
-		}
-
-		// Then check completed operations
-		const completedOperation = this.completedOperations.get(operationId);
-		if (completedOperation) {
-			return completedOperation;
-		}
-
-		// Operation not found in either active or completed
-		return {
-			error: {
-				code: 'OPERATION_NOT_FOUND',
-				message: `Operation ID ${operationId} not found. It may have been completed and removed from history, or the ID may be invalid.`
-			},
-			status: 'not_found'
-		};
-	}
-
-	/**
-	 * Internal logging helper to prefix logs with the operation ID.
-	 * @param {string} operationId - The ID of the operation.
-	 * @param {'info'|'warn'|'error'|'debug'} level - Log level.
-	 * @param {string} message - Log message.
-	 * @param {Object} [meta] - Additional metadata.
-	 */
-	log(operationId, level, message, meta = {}) {
-		const operation = this.operations.get(operationId);
-		// Use the logger instance associated with the operation if available, otherwise console
-		const logger = operation?.log || console;
-		const logFn = logger[level] || logger.log || console.log; // Fallback
-		logFn(`[AsyncOp ${operationId}] ${message}`, meta);
-	}
-
-	// --- Basic Event Emitter ---
-	on(eventName, listener) {
-		if (!this.listeners.has(eventName)) {
-			this.listeners.set(eventName, []);
-		}
-		this.listeners.get(eventName).push(listener);
-	}
-
-	emit(eventName, data) {
-		if (this.listeners.has(eventName)) {
-			this.listeners.get(eventName).forEach((listener) => listener(data));
-		}
-	}
-}
-
-// Export a singleton instance
-const asyncOperationManager = new AsyncOperationManager();
-
-// Export the manager and potentially the class if needed elsewhere
-export { asyncOperationManager, AsyncOperationManager };

mcp-server/src/index.js

@@ -5,7 +5,6 @@ import { fileURLToPath } from 'url';
 import fs from 'fs';
 import logger from './logger.js';
 import { registerTaskMasterTools } from './tools/index.js';
-import { asyncOperationManager } from './core/utils/async-manager.js';
 
 // Load environment variables
 dotenv.config();
@@ -35,9 +34,6 @@ class TaskMasterMCPServer {
 
 		this.server.addResourceTemplate({});
 
-		// Make the manager accessible (e.g., pass it to tool registration)
-		this.asyncManager = asyncOperationManager;
-
 		// Bind methods
 		this.init = this.init.bind(this);
 		this.start = this.start.bind(this);
@@ -88,7 +84,4 @@ class TaskMasterMCPServer {
 	}
 }
 
-// Export the manager from here as well, if needed elsewhere
-export { asyncOperationManager };
-
 export default TaskMasterMCPServer;

mcp-server/src/logger.js

@@ -1,5 +1,6 @@
 import chalk from 'chalk';
 import { isSilentMode } from '../../scripts/modules/utils.js';
+import { getLogLevel } from '../../scripts/modules/config-manager.js';
 
 // Define log levels
 const LOG_LEVELS = {
@@ -10,10 +11,8 @@ const LOG_LEVELS = {
 	success: 4
 };
 
-// Get log level from environment or default to info
-const LOG_LEVEL = process.env.LOG_LEVEL
-	? (LOG_LEVELS[process.env.LOG_LEVEL.toLowerCase()] ?? LOG_LEVELS.info)
-	: LOG_LEVELS.info;
+// Get log level from config manager or default to info
+const LOG_LEVEL = LOG_LEVELS[getLogLevel().toLowerCase()] ?? LOG_LEVELS.info;
 
 /**
  * Logs a message with the specified level

mcp-server/src/tools/add-dependency.js

@@ -7,7 +7,8 @@ import { z } from 'zod';
 import {
 	handleApiResult,
 	createErrorResponse,
-	getProjectRootFromSession
+	getProjectRootFromSession,
+	withNormalizedProjectRoot
 } from './utils.js';
 import { addDependencyDirect } from '../core/task-master-core.js';
 import { findTasksJsonPath } from '../core/utils/path-utils.js';
@@ -35,28 +36,16 @@ export function registerAddDependencyTool(server) {
 				.string()
 				.describe('The directory of the project. Must be an absolute path.')
 		}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
 			try {
 				log.info(
 					`Adding dependency for task ${args.id} to depend on ${args.dependsOn}`
 				);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				// Ensure project root was determined
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
-				// Resolve the path to tasks.json
 				let tasksJsonPath;
 				try {
 					tasksJsonPath = findTasksJsonPath(
-						{ projectRoot: rootFolder, file: args.file },
+						{ projectRoot: args.projectRoot, file: args.file },
 						log
 					);
 				} catch (error) {
@@ -92,6 +81,6 @@ export function registerAddDependencyTool(server) {
 				log.error(`Error in addDependency tool: ${error.message}`);
 				return createErrorResponse(error.message);
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/add-subtask.js

@@ -7,7 +7,7 @@ import { z } from 'zod';
 import {
 	handleApiResult,
 	createErrorResponse,
-	getProjectRootFromSession
+	withNormalizedProjectRoot
 } from './utils.js';
 import { addSubtaskDirect } from '../core/task-master-core.js';
 import { findTasksJsonPath } from '../core/utils/path-utils.js';
@@ -60,24 +60,15 @@ export function registerAddSubtaskTool(server) {
 				.string()
 				.describe('The directory of the project. Must be an absolute path.')
 		}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
 			try {
 				log.info(`Adding subtask with args: ${JSON.stringify(args)}`);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
+				// Use args.projectRoot directly (guaranteed by withNormalizedProjectRoot)
 				let tasksJsonPath;
 				try {
 					tasksJsonPath = findTasksJsonPath(
-						{ projectRoot: rootFolder, file: args.file },
+						{ projectRoot: args.projectRoot, file: args.file },
 						log
 					);
 				} catch (error) {
@@ -113,6 +104,6 @@ export function registerAddSubtaskTool(server) {
 				log.error(`Error in addSubtask tool: ${error.message}`);
 				return createErrorResponse(error.message);
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/add-task.js

@@ -6,10 +6,8 @@
 import { z } from 'zod';
 import {
 	createErrorResponse,
-	createContentResponse,
-	getProjectRootFromSession,
-	executeTaskMasterCommand,
-	handleApiResult
+	handleApiResult,
+	withNormalizedProjectRoot
 } from './utils.js';
 import { addTaskDirect } from '../core/task-master-core.js';
 import { findTasksJsonPath } from '../core/utils/path-utils.js';
@@ -65,26 +63,15 @@ export function registerAddTaskTool(server) {
 				.optional()
 				.describe('Whether to use research capabilities for task creation')
 		}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
 			try {
 				log.info(`Starting add-task with args: ${JSON.stringify(args)}`);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				// Ensure project root was determined
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
-				// Resolve the path to tasks.json
+				// Use args.projectRoot directly (guaranteed by withNormalizedProjectRoot)
 				let tasksJsonPath;
 				try {
 					tasksJsonPath = findTasksJsonPath(
-						{ projectRoot: rootFolder, file: args.file },
+						{ projectRoot: args.projectRoot, file: args.file },
 						log
 					);
 				} catch (error) {
@@ -94,27 +81,29 @@ export function registerAddTaskTool(server) {
 					);
 				}
 
-				// Call the direct function
+				// Call the direct functionP
 				const result = await addTaskDirect(
 					{
-						// Pass the explicitly resolved path
 						tasksJsonPath: tasksJsonPath,
-						// Pass other relevant args
 						prompt: args.prompt,
+						title: args.title,
+						description: args.description,
+						details: args.details,
+						testStrategy: args.testStrategy,
 						dependencies: args.dependencies,
 						priority: args.priority,
-						research: args.research
+						research: args.research,
+						projectRoot: args.projectRoot
 					},
 					log,
 					{ session }
 				);
 
-				// Return the result
 				return handleApiResult(result, log);
 			} catch (error) {
 				log.error(`Error in add-task tool: ${error.message}`);
 				return createErrorResponse(error.message);
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/analyze.js

@@ -4,120 +4,128 @@
  */
 
 import { z } from 'zod';
+import path from 'path';
+import fs from 'fs'; // Import fs for directory check/creation
 import {
 	handleApiResult,
 	createErrorResponse,
-	getProjectRootFromSession
+	withNormalizedProjectRoot
 } from './utils.js';
-import { analyzeTaskComplexityDirect } from '../core/task-master-core.js';
+import { analyzeTaskComplexityDirect } from '../core/task-master-core.js'; // Assuming core functions are exported via task-master-core.js
 import { findTasksJsonPath } from '../core/utils/path-utils.js';
-import path from 'path';
 
 /**
- * Register the analyze tool with the MCP server
+ * Register the analyze_project_complexity tool
  * @param {Object} server - FastMCP server instance
  */
-export function registerAnalyzeTool(server) {
+export function registerAnalyzeProjectComplexityTool(server) {
 	server.addTool({
 		name: 'analyze_project_complexity',
 		description:
-			'Analyze task complexity and generate expansion recommendations',
+			'Analyze task complexity and generate expansion recommendations.',
 		parameters: z.object({
+			threshold: z.coerce // Use coerce for number conversion from string if needed
+				.number()
+				.int()
+				.min(1)
+				.max(10)
+				.optional()
+				.default(5) // Default threshold
+				.describe('Complexity score threshold (1-10) to recommend expansion.'),
+			research: z
+				.boolean()
+				.optional()
+				.default(false)
+				.describe('Use Perplexity AI for research-backed analysis.'),
 			output: z
 				.string()
 				.optional()
 				.describe(
-					'Output file path for the report (default: scripts/task-complexity-report.json)'
-				),
-			model: z
-				.string()
-				.optional()
-				.describe(
-					'LLM model to use for analysis (defaults to configured model)'
-				),
-			threshold: z.coerce
-				.number()
-				.min(1)
-				.max(10)
-				.optional()
-				.describe(
-					'Minimum complexity score to recommend expansion (1-10) (default: 5)'
+					'Output file path relative to project root (default: scripts/task-complexity-report.json).'
 				),
 			file: z
 				.string()
 				.optional()
 				.describe(
-					'Absolute path to the tasks file (default: tasks/tasks.json)'
+					'Path to the tasks file relative to project root (default: tasks/tasks.json).'
 				),
-			research: z
-				.boolean()
-				.optional()
-				.describe('Use Perplexity AI for research-backed complexity analysis'),
 			projectRoot: z
 				.string()
 				.describe('The directory of the project. Must be an absolute path.')
 		}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
+			const toolName = 'analyze_project_complexity'; // Define tool name for logging
 			try {
 				log.info(
-					`Analyzing task complexity with args: ${JSON.stringify(args)}`
+					`Executing ${toolName} tool with args: ${JSON.stringify(args)}`
 				);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
 				let tasksJsonPath;
 				try {
 					tasksJsonPath = findTasksJsonPath(
-						{ projectRoot: rootFolder, file: args.file },
+						{ projectRoot: args.projectRoot, file: args.file },
 						log
 					);
+					log.info(`${toolName}: Resolved tasks path: ${tasksJsonPath}`);
 				} catch (error) {
-					log.error(`Error finding tasks.json: ${error.message}`);
+					log.error(`${toolName}: Error finding tasks.json: ${error.message}`);
 					return createErrorResponse(
-						`Failed to find tasks.json: ${error.message}`
+						`Failed to find tasks.json within project root '${args.projectRoot}': ${error.message}`
 					);
 				}
 
 				const outputPath = args.output
-					? path.resolve(rootFolder, args.output)
-					: path.resolve(rootFolder, 'scripts', 'task-complexity-report.json');
+					? path.resolve(args.projectRoot, args.output)
+					: path.resolve(
+							args.projectRoot,
+							'scripts',
+							'task-complexity-report.json'
+						);
 
+				log.info(`${toolName}: Report output path: ${outputPath}`);
+
+				// Ensure output directory exists
+				const outputDir = path.dirname(outputPath);
+				try {
+					if (!fs.existsSync(outputDir)) {
+						fs.mkdirSync(outputDir, { recursive: true });
+						log.info(`${toolName}: Created output directory: ${outputDir}`);
+					}
+				} catch (dirError) {
+					log.error(
+						`${toolName}: Failed to create output directory ${outputDir}: ${dirError.message}`
+					);
+					return createErrorResponse(
+						`Failed to create output directory: ${dirError.message}`
+					);
+				}
+
+				// 3. Call Direct Function - Pass projectRoot in first arg object
 				const result = await analyzeTaskComplexityDirect(
 					{
 						tasksJsonPath: tasksJsonPath,
 						outputPath: outputPath,
-						model: args.model,
 						threshold: args.threshold,
-						research: args.research
+						research: args.research,
+						projectRoot: args.projectRoot
 					},
 					log,
 					{ session }
 				);
 
-				if (result.success) {
-					log.info(`Task complexity analysis complete: ${result.data.message}`);
-					log.info(
-						`Report summary: ${JSON.stringify(result.data.reportSummary)}`
-					);
-				} else {
-					log.error(
-						`Failed to analyze task complexity: ${result.error.message}`
-					);
-				}
-
+				// 4. Handle Result
+				log.info(
+					`${toolName}: Direct function result: success=${result.success}`
+				);
 				return handleApiResult(result, log, 'Error analyzing task complexity');
 			} catch (error) {
-				log.error(`Error in analyze tool: ${error.message}`);
-				return createErrorResponse(error.message);
+				log.error(
+					`Critical error in ${toolName} tool execute: ${error.message}`
+				);
+				return createErrorResponse(
+					`Internal tool error (${toolName}): ${error.message}`
+				);
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/clear-subtasks.js

@@ -7,7 +7,7 @@ import { z } from 'zod';
 import {
 	handleApiResult,
 	createErrorResponse,
-	getProjectRootFromSession
+	withNormalizedProjectRoot
 } from './utils.js';
 import { clearSubtasksDirect } from '../core/task-master-core.js';
 import { findTasksJsonPath } from '../core/utils/path-utils.js';
@@ -41,26 +41,15 @@ export function registerClearSubtasksTool(server) {
 				message: "Either 'id' or 'all' parameter must be provided",
 				path: ['id', 'all']
 			}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
 			try {
 				log.info(`Clearing subtasks with args: ${JSON.stringify(args)}`);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				// Ensure project root was determined
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
-				// Resolve the path to tasks.json
+				// Use args.projectRoot directly (guaranteed by withNormalizedProjectRoot)
 				let tasksJsonPath;
 				try {
 					tasksJsonPath = findTasksJsonPath(
-						{ projectRoot: rootFolder, file: args.file },
+						{ projectRoot: args.projectRoot, file: args.file },
 						log
 					);
 				} catch (error) {
@@ -72,14 +61,11 @@ export function registerClearSubtasksTool(server) {
 
 				const result = await clearSubtasksDirect(
 					{
-						// Pass the explicitly resolved path
 						tasksJsonPath: tasksJsonPath,
-						// Pass other relevant args
 						id: args.id,
 						all: args.all
 					},
 					log
-					// Remove context object as clearSubtasksDirect likely doesn't need session/reportProgress
 				);
 
 				if (result.success) {
@@ -93,6 +79,6 @@ export function registerClearSubtasksTool(server) {
 				log.error(`Error in clearSubtasks tool: ${error.message}`);
 				return createErrorResponse(error.message);
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/complexity-report.js

@@ -7,7 +7,7 @@ import { z } from 'zod';
 import {
 	handleApiResult,
 	createErrorResponse,
-	getProjectRootFromSession
+	withNormalizedProjectRoot
 } from './utils.js';
 import { complexityReportDirect } from '../core/task-master-core.js';
 import path from 'path';
@@ -31,34 +31,24 @@ export function registerComplexityReportTool(server) {
 				.string()
 				.describe('The directory of the project. Must be an absolute path.')
 		}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
 			try {
 				log.info(
 					`Getting complexity report with args: ${JSON.stringify(args)}`
 				);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				// Ensure project root was determined
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
-				// Resolve the path to the complexity report file
-				// Default to scripts/task-complexity-report.json relative to root
+				// Use args.projectRoot directly (guaranteed by withNormalizedProjectRoot)
 				const reportPath = args.file
-					? path.resolve(rootFolder, args.file)
-					: path.resolve(rootFolder, 'scripts', 'task-complexity-report.json');
+					? path.resolve(args.projectRoot, args.file)
+					: path.resolve(
+							args.projectRoot,
+							'scripts',
+							'task-complexity-report.json'
+						);
 
 				const result = await complexityReportDirect(
 					{
-						// Pass the explicitly resolved path
 						reportPath: reportPath
-						// No other args specific to this tool
 					},
 					log
 				);
@@ -84,6 +74,6 @@ export function registerComplexityReportTool(server) {
 					`Failed to retrieve complexity report: ${error.message}`
 				);
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/expand-all.js

@@ -7,7 +7,7 @@ import { z } from 'zod';
 import {
 	handleApiResult,
 	createErrorResponse,
-	getProjectRootFromSession
+	withNormalizedProjectRoot
 } from './utils.js';
 import { expandAllTasksDirect } from '../core/task-master-core.js';
 import { findTasksJsonPath } from '../core/utils/path-utils.js';
@@ -19,22 +19,27 @@ import { findTasksJsonPath } from '../core/utils/path-utils.js';
 export function registerExpandAllTool(server) {
 	server.addTool({
 		name: 'expand_all',
-		description: 'Expand all pending tasks into subtasks',
+		description:
+			'Expand all pending tasks into subtasks based on complexity or defaults',
 		parameters: z.object({
 			num: z
 				.string()
 				.optional()
-				.describe('Number of subtasks to generate for each task'),
+				.describe(
+					'Target number of subtasks per task (uses complexity/defaults otherwise)'
+				),
 			research: z
 				.boolean()
 				.optional()
 				.describe(
-					'Enable Perplexity AI for research-backed subtask generation'
+					'Enable research-backed subtask generation (e.g., using Perplexity)'
 				),
 			prompt: z
 				.string()
 				.optional()
-				.describe('Additional context to guide subtask generation'),
+				.describe(
+					'Additional context to guide subtask generation for all tasks'
+				),
 			force: z
 				.boolean()
 				.optional()
@@ -45,34 +50,28 @@ export function registerExpandAllTool(server) {
 				.string()
 				.optional()
 				.describe(
-					'Absolute path to the tasks file (default: tasks/tasks.json)'
+					'Absolute path to the tasks file in the /tasks folder inside the project root (default: tasks/tasks.json)'
 				),
 			projectRoot: z
 				.string()
-				.describe('The directory of the project. Must be an absolute path.')
+				.optional()
+				.describe(
+					'Absolute path to the project root directory (derived from session if possible)'
+				)
 		}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
 			try {
-				log.info(`Expanding all tasks with args: ${JSON.stringify(args)}`);
+				log.info(
+					`Tool expand_all execution started with args: ${JSON.stringify(args)}`
+				);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				// Ensure project root was determined
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
-				// Resolve the path to tasks.json
 				let tasksJsonPath;
 				try {
 					tasksJsonPath = findTasksJsonPath(
-						{ projectRoot: rootFolder, file: args.file },
+						{ projectRoot: args.projectRoot, file: args.file },
 						log
 					);
+					log.info(`Resolved tasks.json path: ${tasksJsonPath}`);
 				} catch (error) {
 					log.error(`Error finding tasks.json: ${error.message}`);
 					return createErrorResponse(
@@ -82,31 +81,29 @@ export function registerExpandAllTool(server) {
 
 				const result = await expandAllTasksDirect(
 					{
-						// Pass the explicitly resolved path
 						tasksJsonPath: tasksJsonPath,
-						// Pass other relevant args
 						num: args.num,
 						research: args.research,
 						prompt: args.prompt,
-						force: args.force
+						force: args.force,
+						projectRoot: args.projectRoot
 					},
 					log,
 					{ session }
 				);
 
-				if (result.success) {
-					log.info(`Successfully expanded all tasks: ${result.data.message}`);
-				} else {
-					log.error(
-						`Failed to expand all tasks: ${result.error?.message || 'Unknown error'}`
-					);
-				}
-
 				return handleApiResult(result, log, 'Error expanding all tasks');
 			} catch (error) {
-				log.error(`Error in expand-all tool: ${error.message}`);
-				return createErrorResponse(error.message);
+				log.error(
+					`Unexpected error in expand_all tool execute: ${error.message}`
+				);
+				if (error.stack) {
+					log.error(error.stack);
+				}
+				return createErrorResponse(
+					`An unexpected error occurred: ${error.message}`
+				);
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/expand-task.js

@@ -7,12 +7,10 @@ import { z } from 'zod';
 import {
 	handleApiResult,
 	createErrorResponse,
-	getProjectRootFromSession
+	withNormalizedProjectRoot
 } from './utils.js';
 import { expandTaskDirect } from '../core/task-master-core.js';
 import { findTasksJsonPath } from '../core/utils/path-utils.js';
-import fs from 'fs';
-import path from 'path';
 
 /**
  * Register the expand-task tool with the MCP server
@@ -28,39 +26,36 @@ export function registerExpandTaskTool(server) {
 			research: z
 				.boolean()
 				.optional()
-				.describe('Use Perplexity AI for research-backed generation'),
+				.default(false)
+				.describe('Use research role for generation'),
 			prompt: z
 				.string()
 				.optional()
 				.describe('Additional context for subtask generation'),
-			file: z.string().optional().describe('Absolute path to the tasks file'),
+			file: z
+				.string()
+				.optional()
+				.describe(
+					'Path to the tasks file relative to project root (e.g., tasks/tasks.json)'
+				),
 			projectRoot: z
 				.string()
 				.describe('The directory of the project. Must be an absolute path.'),
-			force: z.boolean().optional().describe('Force the expansion')
+			force: z
+				.boolean()
+				.optional()
+				.default(false)
+				.describe('Force expansion even if subtasks exist')
 		}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
 			try {
 				log.info(`Starting expand-task with args: ${JSON.stringify(args)}`);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				// Ensure project root was determined
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
-				log.info(`Project root resolved to: ${rootFolder}`);
-
-				// Resolve the path to tasks.json using the utility
+				// Use args.projectRoot directly (guaranteed by withNormalizedProjectRoot)
 				let tasksJsonPath;
 				try {
 					tasksJsonPath = findTasksJsonPath(
-						{ projectRoot: rootFolder, file: args.file },
+						{ projectRoot: args.projectRoot, file: args.file },
 						log
 					);
 				} catch (error) {
@@ -70,29 +65,25 @@ export function registerExpandTaskTool(server) {
 					);
 				}
 
-				// Call direct function with only session in the context, not reportProgress
-				// Use the pattern recommended in the MCP guidelines
 				const result = await expandTaskDirect(
 					{
-						// Pass the explicitly resolved path
 						tasksJsonPath: tasksJsonPath,
-						// Pass other relevant args
 						id: args.id,
 						num: args.num,
 						research: args.research,
 						prompt: args.prompt,
-						force: args.force // Need to add force to parameters
+						force: args.force,
+						projectRoot: args.projectRoot
 					},
 					log,
 					{ session }
-				); // Only pass session, NOT reportProgress
+				);
 
-				// Return the result
 				return handleApiResult(result, log, 'Error expanding task');
 			} catch (error) {
-				log.error(`Error in expand task tool: ${error.message}`);
+				log.error(`Error in expand-task tool: ${error.message}`);
 				return createErrorResponse(error.message);
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/fix-dependencies.js

@@ -7,7 +7,7 @@ import { z } from 'zod';
 import {
 	handleApiResult,
 	createErrorResponse,
-	getProjectRootFromSession
+	withNormalizedProjectRoot
 } from './utils.js';
 import { fixDependenciesDirect } from '../core/task-master-core.js';
 import { findTasksJsonPath } from '../core/utils/path-utils.js';
@@ -26,24 +26,15 @@ export function registerFixDependenciesTool(server) {
 				.string()
 				.describe('The directory of the project. Must be an absolute path.')
 		}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
 			try {
 				log.info(`Fixing dependencies with args: ${JSON.stringify(args)}`);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
+				// Use args.projectRoot directly (guaranteed by withNormalizedProjectRoot)
 				let tasksJsonPath;
 				try {
 					tasksJsonPath = findTasksJsonPath(
-						{ projectRoot: rootFolder, file: args.file },
+						{ projectRoot: args.projectRoot, file: args.file },
 						log
 					);
 				} catch (error) {
@@ -71,6 +62,6 @@ export function registerFixDependenciesTool(server) {
 				log.error(`Error in fixDependencies tool: ${error.message}`);
 				return createErrorResponse(error.message);
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/generate.js

@@ -7,7 +7,7 @@ import { z } from 'zod';
 import {
 	handleApiResult,
 	createErrorResponse,
-	getProjectRootFromSession
+	withNormalizedProjectRoot
 } from './utils.js';
 import { generateTaskFilesDirect } from '../core/task-master-core.js';
 import { findTasksJsonPath } from '../core/utils/path-utils.js';
@@ -32,26 +32,15 @@ export function registerGenerateTool(server) {
 				.string()
 				.describe('The directory of the project. Must be an absolute path.')
 		}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
 			try {
 				log.info(`Generating task files with args: ${JSON.stringify(args)}`);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				// Ensure project root was determined
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
-				// Resolve the path to tasks.json
+				// Use args.projectRoot directly (guaranteed by withNormalizedProjectRoot)
 				let tasksJsonPath;
 				try {
 					tasksJsonPath = findTasksJsonPath(
-						{ projectRoot: rootFolder, file: args.file },
+						{ projectRoot: args.projectRoot, file: args.file },
 						log
 					);
 				} catch (error) {
@@ -61,17 +50,14 @@ export function registerGenerateTool(server) {
 					);
 				}
 
-				// Determine output directory: use explicit arg or default to tasks.json directory
 				const outputDir = args.output
-					? path.resolve(rootFolder, args.output) // Resolve relative to root if needed
+					? path.resolve(args.projectRoot, args.output)
 					: path.dirname(tasksJsonPath);
 
 				const result = await generateTaskFilesDirect(
 					{
-						// Pass the explicitly resolved paths
 						tasksJsonPath: tasksJsonPath,
 						outputDir: outputDir
-						// No other args specific to this tool
 					},
 					log
 				);
@@ -89,6 +75,6 @@ export function registerGenerateTool(server) {
 				log.error(`Error in generate tool: ${error.message}`);
 				return createErrorResponse(error.message);
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/get-task.js

@@ -7,7 +7,7 @@ import { z } from 'zod';
 import {
 	handleApiResult,
 	createErrorResponse,
-	getProjectRootFromSession
+	withNormalizedProjectRoot
 } from './utils.js';
 import { showTaskDirect } from '../core/task-master-core.js';
 import { findTasksJsonPath } from '../core/utils/path-utils.js';
@@ -21,8 +21,10 @@ function processTaskResponse(data) {
 	if (!data) return data;
 
 	// If we have the expected structure with task and allTasks
-	if (data.task) {
-		// Return only the task object, removing the allTasks array
+	if (typeof data === 'object' && data !== null && data.id && data.title) {
+		// If the data itself looks like the task object, return it
+		return data;
+	} else if (data.task) {
 		return data.task;
 	}
 
@@ -40,46 +42,37 @@ export function registerShowTaskTool(server) {
 		description: 'Get detailed information about a specific task',
 		parameters: z.object({
 			id: z.string().describe('Task ID to get'),
-			file: z.string().optional().describe('Absolute path to the tasks file'),
+			status: z
+				.string()
+				.optional()
+				.describe("Filter subtasks by status (e.g., 'pending', 'done')"),
+			file: z
+				.string()
+				.optional()
+				.describe('Path to the tasks file relative to project root'),
 			projectRoot: z
 				.string()
-				.describe('The directory of the project. Must be an absolute path.')
+				.optional()
+				.describe(
+					'Absolute path to the project root directory (Optional, usually from session)'
+				)
 		}),
-		execute: async (args, { log, session }) => {
-			// Log the session right at the start of execute
-			log.info(
-				`Session object received in execute: ${JSON.stringify(session)}`
-			); // Use JSON.stringify for better visibility
+		execute: withNormalizedProjectRoot(async (args, { log }) => {
+			const { id, file, status, projectRoot } = args;
 
 			try {
-				log.info(`Getting task details for ID: ${args.id}`);
-
 				log.info(
-					`Session object received in execute: ${JSON.stringify(session)}`
-				); // Use JSON.stringify for better visibility
+					`Getting task details for ID: ${id}${status ? ` (filtering subtasks by status: ${status})` : ''} in root: ${projectRoot}`
+				);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				// Ensure project root was determined
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
-				log.info(`Attempting to use project root: ${rootFolder}`); // Log the final resolved root
-
-				log.info(`Root folder: ${rootFolder}`); // Log the final resolved root
-
-				// Resolve the path to tasks.json
+				// Resolve the path to tasks.json using the NORMALIZED projectRoot from args
 				let tasksJsonPath;
 				try {
 					tasksJsonPath = findTasksJsonPath(
-						{ projectRoot: rootFolder, file: args.file },
+						{ projectRoot: projectRoot, file: file },
 						log
 					);
+					log.info(`Resolved tasks path: ${tasksJsonPath}`);
 				} catch (error) {
 					log.error(`Error finding tasks.json: ${error.message}`);
 					return createErrorResponse(
@@ -87,14 +80,13 @@ export function registerShowTaskTool(server) {
 					);
 				}
 
-				log.info(`Attempting to use tasks file path: ${tasksJsonPath}`);
-
+				// Call the direct function, passing the normalized projectRoot
 				const result = await showTaskDirect(
 					{
-						// Pass the explicitly resolved path
 						tasksJsonPath: tasksJsonPath,
-						// Pass other relevant args
-						id: args.id
+						id: id,
+						status: status,
+						projectRoot: projectRoot
 					},
 					log
 				);
@@ -107,7 +99,7 @@ export function registerShowTaskTool(server) {
 					log.error(`Failed to get task: ${result.error.message}`);
 				}
 
-				// Use our custom processor function to remove allTasks from the response
+				// Use our custom processor function
 				return handleApiResult(
 					result,
 					log,
@@ -115,9 +107,9 @@ export function registerShowTaskTool(server) {
 					processTaskResponse
 				);
 			} catch (error) {
-				log.error(`Error in get-task tool: ${error.message}\n${error.stack}`); // Add stack trace
+				log.error(`Error in get-task tool: ${error.message}\n${error.stack}`);
 				return createErrorResponse(`Failed to get task: ${error.message}`);
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/get-tasks.js

@@ -7,7 +7,7 @@ import { z } from 'zod';
 import {
 	createErrorResponse,
 	handleApiResult,
-	getProjectRootFromSession
+	withNormalizedProjectRoot
 } from './utils.js';
 import { listTasksDirect } from '../core/task-master-core.js';
 import { findTasksJsonPath } from '../core/utils/path-utils.js';
@@ -42,31 +42,19 @@ export function registerListTasksTool(server) {
 				.string()
 				.describe('The directory of the project. Must be an absolute path.')
 		}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
 			try {
 				log.info(`Getting tasks with filters: ${JSON.stringify(args)}`);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				// Ensure project root was determined
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
-				// Resolve the path to tasks.json
+				// Use args.projectRoot directly (guaranteed by withNormalizedProjectRoot)
 				let tasksJsonPath;
 				try {
 					tasksJsonPath = findTasksJsonPath(
-						{ projectRoot: rootFolder, file: args.file },
+						{ projectRoot: args.projectRoot, file: args.file },
 						log
 					);
 				} catch (error) {
 					log.error(`Error finding tasks.json: ${error.message}`);
-					// Use the error message from findTasksJsonPath for better context
 					return createErrorResponse(
 						`Failed to find tasks.json: ${error.message}`
 					);
@@ -89,7 +77,7 @@ export function registerListTasksTool(server) {
 				log.error(`Error getting tasks: ${error.message}`);
 				return createErrorResponse(error.message);
 			}
-		}
+		})
 	});
 }
 

mcp-server/src/tools/index.js

@@ -17,7 +17,7 @@ import { registerExpandTaskTool } from './expand-task.js';
 import { registerAddTaskTool } from './add-task.js';
 import { registerAddSubtaskTool } from './add-subtask.js';
 import { registerRemoveSubtaskTool } from './remove-subtask.js';
-import { registerAnalyzeTool } from './analyze.js';
+import { registerAnalyzeProjectComplexityTool } from './analyze.js';
 import { registerClearSubtasksTool } from './clear-subtasks.js';
 import { registerExpandAllTool } from './expand-all.js';
 import { registerRemoveDependencyTool } from './remove-dependency.js';
@@ -27,39 +27,51 @@ import { registerComplexityReportTool } from './complexity-report.js';
 import { registerAddDependencyTool } from './add-dependency.js';
 import { registerRemoveTaskTool } from './remove-task.js';
 import { registerInitializeProjectTool } from './initialize-project.js';
-import { asyncOperationManager } from '../core/utils/async-manager.js';
+import { registerModelsTool } from './models.js';
 
 /**
  * Register all Task Master tools with the MCP server
  * @param {Object} server - FastMCP server instance
- * @param {asyncOperationManager} asyncManager - The async operation manager instance
  */
-export function registerTaskMasterTools(server, asyncManager) {
+export function registerTaskMasterTools(server) {
 	try {
-		// Register each tool
-		registerListTasksTool(server);
-		registerSetTaskStatusTool(server);
+		// Register each tool in a logical workflow order
+
+		// Group 1: Initialization & Setup
+		registerInitializeProjectTool(server);
+		registerModelsTool(server);
 		registerParsePRDTool(server);
+
+		// Group 2: Task Listing & Viewing
+		registerListTasksTool(server);
+		registerShowTaskTool(server);
+		registerNextTaskTool(server);
+		registerComplexityReportTool(server);
+
+		// Group 3: Task Status & Management
+		registerSetTaskStatusTool(server);
+		registerGenerateTool(server);
+
+		// Group 4: Task Creation & Modification
+		registerAddTaskTool(server);
+		registerAddSubtaskTool(server);
 		registerUpdateTool(server);
 		registerUpdateTaskTool(server);
 		registerUpdateSubtaskTool(server);
-		registerGenerateTool(server);
-		registerShowTaskTool(server);
-		registerNextTaskTool(server);
-		registerExpandTaskTool(server);
-		registerAddTaskTool(server, asyncManager);
-		registerAddSubtaskTool(server);
+		registerRemoveTaskTool(server);
 		registerRemoveSubtaskTool(server);
-		registerAnalyzeTool(server);
 		registerClearSubtasksTool(server);
+
+		// Group 5: Task Analysis & Expansion
+		registerAnalyzeProjectComplexityTool(server);
+		registerExpandTaskTool(server);
 		registerExpandAllTool(server);
+
+		// Group 6: Dependency Management
+		registerAddDependencyTool(server);
 		registerRemoveDependencyTool(server);
 		registerValidateDependenciesTool(server);
 		registerFixDependenciesTool(server);
-		registerComplexityReportTool(server);
-		registerAddDependencyTool(server);
-		registerRemoveTaskTool(server);
-		registerInitializeProjectTool(server);
 	} catch (error) {
 		logger.error(`Error registering Task Master tools: ${error.message}`);
 		throw error;

mcp-server/src/tools/initialize-project.js

@@ -1,8 +1,8 @@
 import { z } from 'zod';
 import {
-	createContentResponse,
 	createErrorResponse,
-	handleApiResult
+	handleApiResult,
+	withNormalizedProjectRoot
 } from './utils.js';
 import { initializeProjectDirect } from '../core/task-master-core.js';
 
@@ -10,32 +10,8 @@ export function registerInitializeProjectTool(server) {
 	server.addTool({
 		name: 'initialize_project',
 		description:
-			"Initializes a new Task Master project structure by calling the core initialization logic. Derives target directory from client session. If project details (name, description, author) are not provided, prompts the user or skips if 'yes' flag is true. DO NOT run without parameters.",
+			'Initializes a new Task Master project structure by calling the core initialization logic. Creates necessary folders and configuration files for Task Master in the current directory.',
 		parameters: z.object({
-			projectName: z
-				.string()
-				.optional()
-				.describe(
-					'The name for the new project. If not provided, prompt the user for it.'
-				),
-			projectDescription: z
-				.string()
-				.optional()
-				.describe(
-					'A brief description for the project. If not provided, prompt the user for it.'
-				),
-			projectVersion: z
-				.string()
-				.optional()
-				.describe(
-					"The initial version for the project (e.g., '0.1.0'). User input not needed unless user requests to override."
-				),
-			authorName: z
-				.string()
-				.optional()
-				.describe(
-					"The author's name. User input not needed unless user requests to override."
-				),
 			skipInstall: z
 				.boolean()
 				.optional()
@@ -47,15 +23,13 @@ export function registerInitializeProjectTool(server) {
 				.boolean()
 				.optional()
 				.default(false)
-				.describe(
-					'Add shell aliases (tm, taskmaster) to shell config file. User input not needed.'
-				),
+				.describe('Add shell aliases (tm, taskmaster) to shell config file.'),
 			yes: z
 				.boolean()
 				.optional()
-				.default(false)
+				.default(true)
 				.describe(
-					"Skip prompts and use default values or provided arguments. Use true if you wish to skip details like the project name, etc. If the project information required for the initialization is not available or provided by the user, prompt if the user wishes to provide them (name, description, author) or skip them. If the user wishes to skip, set the 'yes' flag to true and do not set any other parameters."
+					'Skip prompts and use default values. Always set to true for MCP tools.'
 				),
 			projectRoot: z
 				.string()
@@ -63,19 +37,10 @@ export function registerInitializeProjectTool(server) {
 					'The root directory for the project. ALWAYS SET THIS TO THE PROJECT ROOT DIRECTORY. IF NOT SET, THE TOOL WILL NOT WORK.'
 				)
 		}),
-		execute: async (args, context) => {
+		execute: withNormalizedProjectRoot(async (args, context) => {
 			const { log } = context;
 			const session = context.session;
 
-			log.info(
-				'>>> Full Context Received by Tool:',
-				JSON.stringify(context, null, 2)
-			);
-			log.info(`Context received in tool function: ${context}`);
-			log.info(
-				`Session received in tool function: ${session ? session : 'undefined'}`
-			);
-
 			try {
 				log.info(
 					`Executing initialize_project tool with args: ${JSON.stringify(args)}`
@@ -89,6 +54,6 @@ export function registerInitializeProjectTool(server) {
 				log.error(errorMessage, error);
 				return createErrorResponse(errorMessage, { details: error.stack });
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/models.js

@@ -0,0 +1,79 @@
+/**
+ * models.js
+ * MCP tool for managing AI model configurations
+ */
+
+import { z } from 'zod';
+import {
+	handleApiResult,
+	createErrorResponse,
+	withNormalizedProjectRoot
+} from './utils.js';
+import { modelsDirect } from '../core/task-master-core.js';
+
+/**
+ * Register the models tool with the MCP server
+ * @param {Object} server - FastMCP server instance
+ */
+export function registerModelsTool(server) {
+	server.addTool({
+		name: 'models',
+		description:
+			'Get information about available AI models or set model configurations. Run without arguments to get the current model configuration and API key status for the selected model providers.',
+		parameters: z.object({
+			setMain: z
+				.string()
+				.optional()
+				.describe(
+					'Set the primary model for task generation/updates. Model provider API key is required in the MCP config ENV.'
+				),
+			setResearch: z
+				.string()
+				.optional()
+				.describe(
+					'Set the model for research-backed operations. Model provider API key is required in the MCP config ENV.'
+				),
+			setFallback: z
+				.string()
+				.optional()
+				.describe(
+					'Set the model to use if the primary fails. Model provider API key is required in the MCP config ENV.'
+				),
+			listAvailableModels: z
+				.boolean()
+				.optional()
+				.describe(
+					'List all available models not currently in use. Input/output costs values are in dollars (3 is $3.00).'
+				),
+			projectRoot: z
+				.string()
+				.optional()
+				.describe('The directory of the project. Must be an absolute path.'),
+			openrouter: z
+				.boolean()
+				.optional()
+				.describe('Indicates the set model ID is a custom OpenRouter model.'),
+			ollama: z
+				.boolean()
+				.optional()
+				.describe('Indicates the set model ID is a custom Ollama model.')
+		}),
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
+			try {
+				log.info(`Starting models tool with args: ${JSON.stringify(args)}`);
+
+				// Use args.projectRoot directly (guaranteed by withNormalizedProjectRoot)
+				const result = await modelsDirect(
+					{ ...args, projectRoot: args.projectRoot },
+					log,
+					{ session }
+				);
+
+				return handleApiResult(result, log);
+			} catch (error) {
+				log.error(`Error in models tool: ${error.message}`);
+				return createErrorResponse(error.message);
+			}
+		})
+	});
+}

mcp-server/src/tools/next-task.js

@@ -7,7 +7,7 @@ import { z } from 'zod';
 import {
 	handleApiResult,
 	createErrorResponse,
-	getProjectRootFromSession
+	withNormalizedProjectRoot
 } from './utils.js';
 import { nextTaskDirect } from '../core/task-master-core.js';
 import { findTasksJsonPath } from '../core/utils/path-utils.js';
@@ -27,26 +27,15 @@ export function registerNextTaskTool(server) {
 				.string()
 				.describe('The directory of the project. Must be an absolute path.')
 		}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
 			try {
 				log.info(`Finding next task with args: ${JSON.stringify(args)}`);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				// Ensure project root was determined
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
-				// Resolve the path to tasks.json
+				// Use args.projectRoot directly (guaranteed by withNormalizedProjectRoot)
 				let tasksJsonPath;
 				try {
 					tasksJsonPath = findTasksJsonPath(
-						{ projectRoot: rootFolder, file: args.file },
+						{ projectRoot: args.projectRoot, file: args.file },
 						log
 					);
 				} catch (error) {
@@ -58,9 +47,7 @@ export function registerNextTaskTool(server) {
 
 				const result = await nextTaskDirect(
 					{
-						// Pass the explicitly resolved path
 						tasksJsonPath: tasksJsonPath
-						// No other args specific to this tool
 					},
 					log
 				);
@@ -80,6 +67,6 @@ export function registerNextTaskTool(server) {
 				log.error(`Error in nextTask tool: ${error.message}`);
 				return createErrorResponse(error.message);
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/parse-prd.js

@@ -4,20 +4,16 @@
  */
 
 import { z } from 'zod';
+import path from 'path';
 import {
-	getProjectRootFromSession,
 	handleApiResult,
-	createErrorResponse
+	createErrorResponse,
+	withNormalizedProjectRoot
 } from './utils.js';
 import { parsePRDDirect } from '../core/task-master-core.js';
-import {
-	resolveProjectPaths,
-	findPRDDocumentPath,
-	resolveTasksOutputPath
-} from '../core/utils/path-utils.js';
 
 /**
- * Register the parsePRD tool with the MCP server
+ * Register the parse_prd tool
  * @param {Object} server - FastMCP server instance
  */
 export function registerParsePRDTool(server) {
@@ -46,65 +42,50 @@ export function registerParsePRDTool(server) {
 			force: z
 				.boolean()
 				.optional()
-				.describe('Allow overwriting an existing tasks.json file.'),
+				.default(false)
+				.describe('Overwrite existing output file without prompting.'),
+			append: z
+				.boolean()
+				.optional()
+				.default(false)
+				.describe('Append generated tasks to existing file.'),
 			projectRoot: z
 				.string()
-				.describe('The directory of the project. Must be absolute path.')
+				.describe('The directory of the project. Must be an absolute path.')
 		}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
+			const toolName = 'parse_prd';
 			try {
-				log.info(`Parsing PRD with args: ${JSON.stringify(args)}`);
-
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
-				// Resolve input (PRD) and output (tasks.json) paths using the utility
-				const { projectRoot, prdPath, tasksJsonPath } = resolveProjectPaths(
-					rootFolder,
-					args,
-					log
+				log.info(
+					`Executing ${toolName} tool with args: ${JSON.stringify(args)}`
 				);
 
-				// Check if PRD path was found (resolveProjectPaths returns null if not found and not provided)
-				if (!prdPath) {
-					return createErrorResponse(
-						'No PRD document found or provided. Please ensure a PRD file exists (e.g., PRD.md) or provide a valid input file path.'
-					);
-				}
-
-				// Call the direct function with fully resolved paths
+				// Call Direct Function - Pass relevant args including projectRoot
 				const result = await parsePRDDirect(
 					{
-						projectRoot: projectRoot,
-						input: prdPath,
-						output: tasksJsonPath,
+						input: args.input,
+						output: args.output,
 						numTasks: args.numTasks,
-						force: args.force
+						force: args.force,
+						append: args.append,
+						projectRoot: args.projectRoot
 					},
 					log,
 					{ session }
 				);
 
-				if (result.success) {
-					log.info(`Successfully parsed PRD: ${result.data.message}`);
-				} else {
-					log.error(
-						`Failed to parse PRD: ${result.error?.message || 'Unknown error'}`
-					);
-				}
-
+				log.info(
+					`${toolName}: Direct function result: success=${result.success}`
+				);
 				return handleApiResult(result, log, 'Error parsing PRD');
 			} catch (error) {
-				log.error(`Error in parse-prd tool: ${error.message}`);
-				return createErrorResponse(error.message);
+				log.error(
+					`Critical error in ${toolName} tool execute: ${error.message}`
+				);
+				return createErrorResponse(
+					`Internal tool error (${toolName}): ${error.message}`
+				);
 			}
-		}
+		})
 	});
 }

mcp-server/src/tools/remove-dependency.js

@@ -7,7 +7,7 @@ import { z } from 'zod';
 import {
 	handleApiResult,
 	createErrorResponse,
-	getProjectRootFromSession
+	withNormalizedProjectRoot
 } from './utils.js';
 import { removeDependencyDirect } from '../core/task-master-core.js';
 import { findTasksJsonPath } from '../core/utils/path-utils.js';
@@ -33,28 +33,17 @@ export function registerRemoveDependencyTool(server) {
 				.string()
 				.describe('The directory of the project. Must be an absolute path.')
 		}),
-		execute: async (args, { log, session }) => {
+		execute: withNormalizedProjectRoot(async (args, { log, session }) => {
 			try {
 				log.info(
 					`Removing dependency for task ${args.id} from ${args.dependsOn} with args: ${JSON.stringify(args)}`
 				);
 
-				// Get project root from args or session
-				const rootFolder =
-					args.projectRoot || getProjectRootFromSession(session, log);
-
-				// Ensure project root was determined
-				if (!rootFolder) {
-					return createErrorResponse(
-						'Could not determine project root. Please provide it explicitly or ensure your session contains valid root information.'
-					);
-				}
-
-				// Resolve the path to tasks.json
+				// Use args.projectRoot directly (guaranteed by withNormalizedProjectRoot)
 				let tasksJsonPath;
 				try {
 					tasksJsonPath = findTasksJsonPath(
-						{ projectRoot: rootFolder, file: args.file },
+						{ projectRoot: args.projectRoot, file: args.file },
 						log
 					);
 				} catch (error) {
@@ -66,9 +55,7 @@ export function registerRemoveDependencyTool(server) {
 
 				const result = await removeDependencyDirect(
 					{
-						// Pass the explicitly resolved path
 						tasksJsonPath: tasksJsonPath,
-						// Pass other relevant args
 						id: args.id,
 						dependsOn: args.dependsOn
 					},
@@ -86,6 +73,6 @@ export function registerRemoveDependencyTool(server) {
 				log.error(`Error in removeDependency tool: ${error.message}`);
 				return createErrorResponse(error.message);
 			}
-		}
+		})
 	});
 }