security: tighten permissions and add debug step

- Change issues permission from write to read (least privilege)
- Add debug step to show generated metrics files before parsing
- Helps troubleshoot empty/missing files without failing the job
- Updated file list to match current pr_created_metrics.md structure

.changeset/easy-deer-heal.md

@@ -1,5 +0,0 @@
----
-"task-master-ai": minor
----
-
-Testing one more pre-release iteration

.changeset/moody-oranges-slide.md

@@ -1,5 +0,0 @@
----
-"task-master-ai": minor
----
-
-Test out the RC

.changeset/odd-otters-tan.md

@@ -1,5 +0,0 @@
----
-"@tm/cli": minor
----
-
-testing this stuff out to see how the release candidate works with monorepo

.changeset/pre.json

@@ -1,5 +1,5 @@
 {
-  "mode": "pre",
+  "mode": "exit",
   "tag": "rc",
   "initialVersions": {
     "task-master-ai": "0.26.0",

.changeset/pretty-planes-cross.md

@@ -2,7 +2,7 @@
 "task-master-ai": minor
 ---
 
-Add grok-cli as a provider. You can now use Grok models with Task Master by setting the `GROK_CLI_API_KEY` environment variable.
+Add grok-cli as a provider with full codebase context support. You can now use Grok models (grok-2, grok-3, grok-4, etc.) with Task Master for AI operations that have access to your entire codebase context, enabling more informed task generation and PRD parsing.
 
 ## Setup Instructions
 
@@ -20,11 +20,17 @@ Add grok-cli as a provider. You can now use Grok models with Task Master by sett
    task-master models --set-fallback grok-beta
    ```
 
-## Available Models
-- `grok-beta` - Latest Grok model
-- `grok-vision-beta` - Grok with vision capabilities
+## Key Features
+- **Full codebase context**: Grok models can analyze your entire project when generating tasks or parsing PRDs
+- **xAI model access**: Support for latest Grok models (grok-2, grok-3, grok-4, etc.)
+- **Code-aware task generation**: Create more accurate and contextual tasks based on your actual codebase
+- **Intelligent PRD parsing**: Parse requirements with understanding of your existing code structure
 
-The Grok CLI provider integrates with xAI's Grok models and can also use the local Grok CLI configuration file (`~/.grok/user-settings.json`) if available.
+## Available Models
+- `grok-beta` - Latest Grok model with codebase context
+- `grok-vision-beta` - Grok with vision capabilities and codebase context
+
+The Grok CLI provider integrates with xAI's Grok models via grok-cli and can also use the local Grok CLI configuration file (`~/.grok/user-settings.json`) if available.
 
 ## Credits
 Built using the [grok-cli](https://github.com/superagent-ai/grok-cli) by Superagent AI for seamless integration with xAI's Grok models.

.changeset/ready-plums-worry.md

@@ -0,0 +1,8 @@
+---
+"task-master-ai": minor
+---
+
+Improve taskmaster ai provider defaults
+
+- moving from main anthropic 3.7 to anthropic sonnet 4
+- moving from fallback anthropic 3.5 to anthropic 3.7

.github/scripts/parse-metrics.mjs

@@ -0,0 +1,157 @@
+#!/usr/bin/env node
+
+import { readFileSync, existsSync, writeFileSync } from 'fs';
+
+function parseMetricsTable(content, metricName) {
+	const lines = content.split('\n');
+
+	for (let i = 0; i < lines.length; i++) {
+		const line = lines[i].trim();
+		// Match a markdown table row like: | Metric Name | value | ...
+		const safeName = metricName.replace(/[.*+?^${}()|[\]\\]/g, '\\$&');
+		const re = new RegExp(`^\\|\\s*${safeName}\\s*\\|\\s*([^|]+)\\|?`);
+		const match = line.match(re);
+		if (match) {
+			return match[1].trim() || 'N/A';
+		}
+	}
+	return 'N/A';
+}
+
+function parseCountMetric(content, metricName) {
+	const result = parseMetricsTable(content, metricName);
+	// Extract number from string, handling commas and spaces
+	const numberMatch = result.toString().match(/[\d,]+/);
+	if (numberMatch) {
+		const number = parseInt(numberMatch[0].replace(/,/g, ''));
+		return isNaN(number) ? 0 : number;
+	}
+	return 0;
+}
+
+function main() {
+	const metrics = {
+		issues_created: 0,
+		issues_closed: 0,
+		prs_created: 0,
+		prs_merged: 0,
+		issue_avg_first_response: 'N/A',
+		issue_avg_time_to_close: 'N/A',
+		pr_avg_first_response: 'N/A',
+		pr_avg_merge_time: 'N/A'
+	};
+
+	// Parse issue metrics
+	if (existsSync('issue_metrics.md')) {
+		console.log('📄 Found issue_metrics.md, parsing...');
+		const issueContent = readFileSync('issue_metrics.md', 'utf8');
+
+		metrics.issues_created = parseCountMetric(
+			issueContent,
+			'Total number of items created'
+		);
+		metrics.issues_closed = parseCountMetric(
+			issueContent,
+			'Number of items closed'
+		);
+		metrics.issue_avg_first_response = parseMetricsTable(
+			issueContent,
+			'Time to first response'
+		);
+		metrics.issue_avg_time_to_close = parseMetricsTable(
+			issueContent,
+			'Time to close'
+		);
+	} else {
+		console.warn('[parse-metrics] issue_metrics.md not found; using defaults.');
+	}
+
+	// Parse PR created metrics
+	if (existsSync('pr_created_metrics.md')) {
+		console.log('📄 Found pr_created_metrics.md, parsing...');
+		const prCreatedContent = readFileSync('pr_created_metrics.md', 'utf8');
+
+		metrics.prs_created = parseCountMetric(
+			prCreatedContent,
+			'Total number of items created'
+		);
+		metrics.pr_avg_first_response = parseMetricsTable(
+			prCreatedContent,
+			'Time to first response'
+		);
+	} else {
+		console.warn(
+			'[parse-metrics] pr_created_metrics.md not found; using defaults.'
+		);
+	}
+
+	// Parse PR merged metrics (for more accurate merge data)
+	if (existsSync('pr_merged_metrics.md')) {
+		console.log('📄 Found pr_merged_metrics.md, parsing...');
+		const prMergedContent = readFileSync('pr_merged_metrics.md', 'utf8');
+
+		metrics.prs_merged = parseCountMetric(
+			prMergedContent,
+			'Total number of items created'
+		);
+		// For merged PRs, "Time to close" is actually time to merge
+		metrics.pr_avg_merge_time = parseMetricsTable(
+			prMergedContent,
+			'Time to close'
+		);
+	} else {
+		console.warn(
+			'[parse-metrics] pr_merged_metrics.md not found; falling back to pr_metrics.md.'
+		);
+		// Fallback: try old pr_metrics.md if it exists
+		if (existsSync('pr_metrics.md')) {
+			console.log('📄 Falling back to pr_metrics.md...');
+			const prContent = readFileSync('pr_metrics.md', 'utf8');
+
+			const mergedCount = parseCountMetric(prContent, 'Number of items merged');
+			metrics.prs_merged =
+				mergedCount || parseCountMetric(prContent, 'Number of items closed');
+
+			const maybeMergeTime = parseMetricsTable(
+				prContent,
+				'Average time to merge'
+			);
+			metrics.pr_avg_merge_time =
+				maybeMergeTime !== 'N/A'
+					? maybeMergeTime
+					: parseMetricsTable(prContent, 'Time to close');
+		} else {
+			console.warn('[parse-metrics] pr_metrics.md not found; using defaults.');
+		}
+	}
+
+	// Output for GitHub Actions
+	const output = Object.entries(metrics)
+		.map(([key, value]) => `${key}=${value}`)
+		.join('\n');
+
+	// Always output to stdout for debugging
+	console.log('\n=== FINAL METRICS ===');
+	Object.entries(metrics).forEach(([key, value]) => {
+		console.log(`${key}: ${value}`);
+	});
+
+	// Write to GITHUB_OUTPUT if in GitHub Actions
+	if (process.env.GITHUB_OUTPUT) {
+		try {
+			writeFileSync(process.env.GITHUB_OUTPUT, output + '\n', { flag: 'a' });
+			console.log(
+				`\nSuccessfully wrote metrics to ${process.env.GITHUB_OUTPUT}`
+			);
+		} catch (error) {
+			console.error(`Failed to write to GITHUB_OUTPUT: ${error.message}`);
+			process.exit(1);
+		}
+	} else {
+		console.log(
+			'\nNo GITHUB_OUTPUT environment variable found, skipping file write'
+		);
+	}
+}
+
+main();

.github/workflows/weekly-metrics-discord.yml

@@ -8,7 +8,7 @@ on:
 
 permissions:
   contents: read
-  issues: write
+  issues: read
   pull-requests: read
 
 jobs:
@@ -17,15 +17,25 @@ jobs:
     env:
       DISCORD_WEBHOOK: ${{ secrets.DISCORD_METRICS_WEBHOOK }}
     steps:
-      - name: Get dates for last week
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+
+      - name: Get dates for last 14 days
         run: |
-          # Last 7 days
-          first_day=$(date -d "7 days ago" +%Y-%m-%d)
+          set -Eeuo pipefail
+          # Last 14 days
+          first_day=$(date -d "14 days ago" +%Y-%m-%d)
           last_day=$(date +%Y-%m-%d)
 
           echo "first_day=$first_day" >> $GITHUB_ENV
           echo "last_day=$last_day" >> $GITHUB_ENV
           echo "week_of=$(date -d '7 days ago' +'Week of %B %d, %Y')" >> $GITHUB_ENV
+          echo "date_range=Past 14 days ($first_day to $last_day)" >> $GITHUB_ENV
 
       - name: Generate issue metrics
         uses: github/issue-metrics@v3
@@ -34,40 +44,39 @@ jobs:
           SEARCH_QUERY: "repo:${{ github.repository }} is:issue created:${{ env.first_day }}..${{ env.last_day }}"
           HIDE_TIME_TO_ANSWER: true
           HIDE_LABEL_METRICS: false
+          OUTPUT_FILE: issue_metrics.md
 
-      - name: Generate PR metrics
+      - name: Generate PR created metrics
         uses: github/issue-metrics@v3
         env:
           GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
           SEARCH_QUERY: "repo:${{ github.repository }} is:pr created:${{ env.first_day }}..${{ env.last_day }}"
-          OUTPUT_FILE: pr_metrics.md
+          OUTPUT_FILE: pr_created_metrics.md
+
+      - name: Generate PR merged metrics
+        uses: github/issue-metrics@v3
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          SEARCH_QUERY: "repo:${{ github.repository }} is:pr is:merged merged:${{ env.first_day }}..${{ env.last_day }}"
+          OUTPUT_FILE: pr_merged_metrics.md
+
+      - name: Debug generated metrics
+        run: |
+          set -Eeuo pipefail
+          echo "Listing markdown files in workspace:"
+          ls -la *.md || true
+          for f in issue_metrics.md pr_created_metrics.md pr_merged_metrics.md; do
+            if [ -f "$f" ]; then
+              echo "== $f (first 10 lines) =="
+              head -n 10 "$f"
+            else
+              echo "Missing $f"
+            fi
+          done
 
       - name: Parse metrics
         id: metrics
-        run: |
-          # Parse the metrics from the generated markdown files
-          if [ -f "issue_metrics.md" ]; then
-            # Extract key metrics using grep/awk
-            AVG_TIME_TO_FIRST_RESPONSE=$(grep -A 1 "Average time to first response" issue_metrics.md | tail -1 | xargs || echo "N/A")
-            AVG_TIME_TO_CLOSE=$(grep -A 1 "Average time to close" issue_metrics.md | tail -1 | xargs || echo "N/A")
-            NUM_ISSUES_CREATED=$(grep -oP '\d+(?= issues created)' issue_metrics.md || echo "0")
-            NUM_ISSUES_CLOSED=$(grep -oP '\d+(?= issues closed)' issue_metrics.md || echo "0")
-          fi
-
-          if [ -f "pr_metrics.md" ]; then
-            PR_AVG_TIME_TO_MERGE=$(grep -A 1 "Average time to close" pr_metrics.md | tail -1 | xargs || echo "N/A")
-            NUM_PRS_CREATED=$(grep -oP '\d+(?= pull requests created)' pr_metrics.md || echo "0")
-            NUM_PRS_MERGED=$(grep -oP '\d+(?= pull requests closed)' pr_metrics.md || echo "0")
-          fi
-
-          # Set outputs for Discord action
-          echo "issues_created=${NUM_ISSUES_CREATED:-0}" >> $GITHUB_OUTPUT
-          echo "issues_closed=${NUM_ISSUES_CLOSED:-0}" >> $GITHUB_OUTPUT
-          echo "prs_created=${NUM_PRS_CREATED:-0}" >> $GITHUB_OUTPUT
-          echo "prs_merged=${NUM_PRS_MERGED:-0}" >> $GITHUB_OUTPUT
-          echo "avg_first_response=${AVG_TIME_TO_FIRST_RESPONSE:-N/A}" >> $GITHUB_OUTPUT
-          echo "avg_time_to_close=${AVG_TIME_TO_CLOSE:-N/A}" >> $GITHUB_OUTPUT
-          echo "pr_avg_merge_time=${PR_AVG_TIME_TO_MERGE:-N/A}" >> $GITHUB_OUTPUT
+        run: node .github/scripts/parse-metrics.mjs
 
       - name: Send to Discord
         uses: sarisia/actions-status-discord@v1
@@ -78,19 +87,22 @@ jobs:
           title: "📊 Weekly Metrics Report"
           description: |
             **${{ env.week_of }}**
-            
+            *${{ env.date_range }}*
+
             **🎯 Issues**
             • Created: ${{ steps.metrics.outputs.issues_created }}
             • Closed: ${{ steps.metrics.outputs.issues_closed }}
-            
+            • Avg Response Time: ${{ steps.metrics.outputs.issue_avg_first_response }}
+            • Avg Time to Close: ${{ steps.metrics.outputs.issue_avg_time_to_close }}
+
             **🔀 Pull Requests**
             • Created: ${{ steps.metrics.outputs.prs_created }}
             • Merged: ${{ steps.metrics.outputs.prs_merged }}
-            
-            **⏱️ Response Times**
-            • First Response: ${{ steps.metrics.outputs.avg_first_response }}
-            • Time to Close: ${{ steps.metrics.outputs.avg_time_to_close }}
-            • PR Merge Time: ${{ steps.metrics.outputs.pr_avg_merge_time }}
+            • Avg Response Time: ${{ steps.metrics.outputs.pr_avg_first_response }}
+            • Avg Time to Merge: ${{ steps.metrics.outputs.pr_avg_merge_time }}
+
+            **📈 Visual Analytics**
+            https://repobeats.axiom.co/api/embed/b439f28f0ab5bd7a2da19505355693cd2c55bfd4.svg
           color: 0x58AFFF
           username: Task Master Metrics Bot
           avatar_url: https://raw.githubusercontent.com/eyaltoledano/claude-task-master/main/images/logo.png

.taskmaster/config.json

@@ -1,9 +1,9 @@
 {
 	"models": {
 		"main": {
-			"provider": "grok-cli",
-			"modelId": "grok-4-latest",
-			"maxTokens": 131072,
+			"provider": "anthropic",
+			"modelId": "claude-sonnet-4-20250514",
+			"maxTokens": 64000,
 			"temperature": 0.2
 		},
 		"research": {
@@ -14,8 +14,8 @@
 		},
 		"fallback": {
 			"provider": "anthropic",
-			"modelId": "claude-sonnet-4-20250514",
-			"maxTokens": 64000,
+			"modelId": "claude-3-7-sonnet-20250219",
+			"maxTokens": 120000,
 			"temperature": 0.2
 		}
 	},

apps/docs/best-practices/configuration-advanced.mdx

@@ -75,7 +75,6 @@ Taskmaster uses two primary methods for configuration:
   - `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.
-  - `GROK_CLI_API_KEY`: Your Grok API key from console.x.ai.
 - **Optional Endpoint Overrides:**
   - **Per-role `baseURL` in `.taskmasterconfig`:** You can add a `baseURL` property to any model role (`main`, `research`, `fallback`) to override the default API endpoint for that provider. If omitted, the provider's standard endpoint is used.
   - **Environment Variable Overrides (`<PROVIDER>_BASE_URL`):** For greater flexibility, especially with third-party services, you can set an environment variable like `OPENAI_BASE_URL` or `MISTRAL_BASE_URL`. This will override any `baseURL` set in the configuration file for that provider. This is the recommended way to connect to OpenAI-compatible APIs.
@@ -317,68 +316,4 @@ Azure OpenAI provides enterprise-grade OpenAI models through Microsoft's Azure c
    - Confirm the model is deployed in your Azure OpenAI resource
    - Verify the deployment name matches your configuration exactly (case-sensitive)
    - Ensure the model deployment is in a "Succeeded" state in Azure OpenAI Studio
-   - Ensure youre not getting rate limited by `maxTokens` maintain appropriate Tokens per Minute Rate Limit (TPM) in your deployment.
-
-### Grok AI Configuration
-
-Grok AI provides access to xAI's Grok models with enhanced reasoning capabilities and requires minimal configuration:
-
-1. **Prerequisites**:
-   - An xAI account with API access
-   - Grok API key from [console.x.ai](https://console.x.ai)
-
-2. **Authentication**:
-   - Set the `GROK_CLI_API_KEY` environment variable with your Grok API key
-
-3. **Available Models**:
-   - `grok-beta`: Latest Grok model with advanced reasoning
-   - `grok-vision-beta`: Grok with vision capabilities for image analysis
-
-4. **Configuration Example**:
-   ```json
-   // In .taskmaster/config.json
-   {
-     "models": {
-       "main": {
-         "provider": "grok-cli",
-         "modelId": "grok-beta",
-         "maxTokens": 131072,
-         "temperature": 0.3
-       },
-       "research": {
-         "provider": "grok-cli",
-         "modelId": "grok-vision-beta",
-         "maxTokens": 131072,
-         "temperature": 0.1
-       }
-     }
-   }
-   ```
-
-5. **Environment Variables**:
-   ```bash
-   # In .env file
-   GROK_CLI_API_KEY=your-grok-api-key-here
-   ```
-
-6. **Setup Commands**:
-   ```bash
-   # Set Grok as your main model
-   task-master models --set-main grok-beta
-   
-   # Set Grok as your research model
-   task-master models --set-research grok-beta
-   
-   # Set Grok as your fallback model
-   task-master models --set-fallback grok-beta
-   ```
-
-7. **Integration Features**:
-   - **Local Configuration Support**: The Grok CLI provider can use your local Grok CLI configuration file (`~/.grok/user-settings.json`) if available
-   - **Full Token Capacity**: Supports Grok's full 131K token capacity for large context operations
-   - **Built on Grok CLI**: Uses the [grok-cli](https://github.com/superagent-ai/grok-cli) by Superagent AI for reliable integration
-
-8. **Troubleshooting**:
-   - **API Key Issues**: Verify your `GROK_CLI_API_KEY` is correctly set and valid
-   - **Model Availability**: Ensure you have access to the specified Grok model variant
-   - **Rate Limits**: Grok models have generous rate limits, but large contexts may take longer to process
+   - Ensure youre not getting rate limited by `maxTokens` maintain appropriate Tokens per Minute Rate Limit (TPM) in your deployment.

apps/docs/capabilities/cli-root-commands.mdx

@@ -38,39 +38,6 @@ sidebarTitle: "CLI Commands"
     ```
   </Accordion>
 
-  <Accordion title="Start Working on a Task">
-    ```bash
-    # Start working on a specific task with Claude Code
-    task-master start <id>
-    
-    # Start the next available task automatically
-    task-master start
-    
-    # Show what would be executed without launching Claude Code
-    task-master start <id> --dry-run
-    
-    # Force start even if another task is in-progress
-    task-master start <id> --force
-    
-    # Don't automatically update task status to in-progress
-    task-master start <id> --no-status-update
-    
-    # Specify project root directory
-    task-master start <id> --project /path/to/project
-    
-    # Get results in JSON format
-    task-master start <id> --format json
-    ```
-    
-    The `start` command automatically launches Claude Code with comprehensive context about the task, including:
-    - Task details and requirements
-    - Implementation guidelines
-    - Related subtasks and dependencies
-    - Project-specific context
-    
-    When no task ID is provided, it automatically finds and starts the next available task based on dependencies and status.
-  </Accordion>
-
   <Accordion title="Show Specific Task">
     ```bash
     # Show details of a specific task

apps/docs/getting-started/quick-start/execute-quick.mdx

@@ -16,24 +16,6 @@ Alternatively you can use the CLI to show the next task
 task-master next
 ```
 
-### Quick Start with `task-master start`
-
-For immediate task execution, you can use the new `start` command to automatically launch Claude Code with full task context:
-
-```bash
-# Start the next available task automatically
-task-master start
-
-# Or start a specific task
-task-master start 1.2
-```
-
-This command will:
-- Find the next available task (if no ID is provided)
-- Update the task status to "in-progress"  
-- Launch Claude Code with comprehensive task context
-- Provide all necessary implementation details and project context
-
 ## Discuss Task
 When you know what task to work on next you can then start chatting with the agent to make sure it understands the plan of action.
 

apps/docs/whats-new.mdx

@@ -3,70 +3,4 @@ title: "What's New"
 sidebarTitle: "What's New"
 ---
 
-## Latest Features (January 2025)
-
-### 🚀 New `task-master start` Command
-
-**Automated Task Execution with Claude Code Integration**
-
-The new `start` command revolutionizes your development workflow by automatically launching Claude Code with comprehensive task context:
-
-```bash
-# Start a specific task
-task-master start 1.2
-
-# Start the next available task automatically
-task-master start
-
-# Preview what would be executed without launching Claude Code
-task-master start 1.2 --dry-run
-```
-
-**Key Features:**
-- **Automatic Task Discovery** - When no ID is provided, finds the next available task based on dependencies and status
-- **Rich Context Injection** - Provides Claude Code with task details, requirements, subtasks, and project context
-- **Status Management** - Automatically updates task status to "in-progress" when starting
-- **Flexible Options** - Support for dry-run, force mode, custom project paths, and JSON output
-
-### 🤖 Grok AI Provider Support
-
-**Enhanced AI Model Options**
-
-Task Master now supports xAI's Grok models with full 131K token capacity:
-
-```bash
-# Configure Grok as your main model
-task-master models --set-main grok-beta
-
-# Use Grok with vision capabilities
-task-master models --set-research grok-vision-beta
-```
-
-**Setup:**
-1. Get your API key from [console.x.ai](https://console.x.ai)
-2. Set `GROK_CLI_API_KEY` environment variable
-3. Configure using `task-master models --setup`
-
-**Available Models:**
-- `grok-beta` - Latest Grok model with advanced reasoning
-- `grok-vision-beta` - Grok with vision capabilities
-
-### 📱 VS Code Extension "Start Task" Button
-
-**Seamless VS Code Integration**
-
-The Task Master VS Code extension now includes a "Start Task" button for one-click task execution:
-
-- **Direct Integration** - Launch Claude Code directly from task cards in VS Code
-- **No Terminal Switching** - Automatic terminal management and command execution
-- **Full Context** - Same rich context injection as the CLI command
-- **Visual Workflow** - Seamless transition from task planning to implementation
-
-### 🔧 Technical Improvements
-
-- **TypeScript Migration** - Core components now use TypeScript for better type safety
-- **Model Configuration Updates** - Upgraded fallback model to Claude Sonnet 4
-- **Token Capacity Fixes** - Grok models now properly support their full 131K token capacity
-- **Enhanced Error Handling** - Improved error messages and debugging capabilities
-
 An easy way to see the latest releases

apps/extension/README.md

@@ -22,7 +22,6 @@ Taskmaster AI is an intelligent task management system designed for AI-assisted
 ![Kanban Board](assets/screenshots/kanban-board.png)
 
 ### 🤖 **AI-Powered Features**
-- **One-Click Task Start** - Launch Claude Code directly from task cards with full context
 - **Task Content Generation** - Regenerate task descriptions using AI
 - **Smart Task Updates** - Append findings and progress notes automatically
 - **MCP Integration** - Seamless connection to Taskmaster AI via Model Context Protocol
@@ -84,7 +83,6 @@ The extension automatically handles the Taskmaster MCP server connection:
 | **View Kanban Board** | `Ctrl/Cmd + Shift + P` → "Taskmaster: Show Board" |
 | **Change Task Status** | Drag task card to different column |
 | **View Task Details** | Click on any task card |
-| **Start Working on Task** | Click "Start Task" button to launch Claude Code automatically |
 | **Edit Task Content** | Click task → Use edit buttons in details panel |
 | **Add Subtasks** | Click the + button on parent task cards |
 | **Use AI Features** | Open task details → Click AI action buttons |

packages/tm-core/src/config/config-manager.spec.ts

@@ -5,6 +5,7 @@
 
 import { describe, it, expect, beforeEach, vi, afterEach } from 'vitest';
 import { ConfigManager } from './config-manager.js';
+import { DEFAULT_CONFIG_VALUES } from '../interfaces/configuration.interface.js';
 import { ConfigLoader } from './services/config-loader.service.js';
 import { ConfigMerger } from './services/config-merger.service.js';
 import { RuntimeStateManager } from './services/runtime-state-manager.service.js';
@@ -69,8 +70,8 @@ describe('ConfigManager', () => {
 				({
 					loadState: vi.fn().mockResolvedValue({ activeTag: 'master' }),
 					saveState: vi.fn().mockResolvedValue(undefined),
-					getActiveTag: vi.fn().mockReturnValue('master'),
-					setActiveTag: vi.fn().mockResolvedValue(undefined),
+					getCurrentTag: vi.fn().mockReturnValue('master'),
+					setCurrentTag: vi.fn().mockResolvedValue(undefined),
 					getState: vi.fn().mockReturnValue({ activeTag: 'master' }),
 					updateMetadata: vi.fn().mockResolvedValue(undefined),
 					clearState: vi.fn().mockResolvedValue(undefined)
@@ -227,8 +228,8 @@ describe('ConfigManager', () => {
 
 			const models = manager.getModelConfig();
 			expect(models).toEqual({
-				main: 'claude-3-5-sonnet-20241022',
-				fallback: 'gpt-4o-mini'
+				main: DEFAULT_CONFIG_VALUES.MODELS.MAIN,
+				fallback: DEFAULT_CONFIG_VALUES.MODELS.FALLBACK
 			});
 		});
 
@@ -281,7 +282,7 @@ describe('ConfigManager', () => {
 			await manager.setActiveTag('feature-branch');
 
 			const stateManager = (manager as any).stateManager;
-			expect(stateManager.setActiveTag).toHaveBeenCalledWith('feature-branch');
+			expect(stateManager.setCurrentTag).toHaveBeenCalledWith('feature-branch');
 		});
 	});
 

packages/tm-core/src/config/config-manager.ts

@@ -10,6 +10,7 @@ import type {
 	PartialConfiguration,
 	RuntimeStorageConfig
 } from '../interfaces/configuration.interface.js';
+import { DEFAULT_CONFIG_VALUES as DEFAULTS } from '../interfaces/configuration.interface.js';
 import { ConfigLoader } from './services/config-loader.service.js';
 import {
 	ConfigMerger,
@@ -167,8 +168,8 @@ export class ConfigManager {
 	getModelConfig() {
 		return (
 			this.config.models || {
-				main: 'claude-3-5-sonnet-20241022',
-				fallback: 'gpt-4o-mini'
+				main: DEFAULTS.MODELS.MAIN,
+				fallback: DEFAULTS.MODELS.FALLBACK
 			}
 		);
 	}

packages/tm-core/src/interfaces/configuration.interface.ts

@@ -399,8 +399,8 @@ export interface IConfigurationManager {
  */
 export const DEFAULT_CONFIG_VALUES = {
 	MODELS: {
-		MAIN: 'claude-3-5-sonnet-20241022',
-		FALLBACK: 'gpt-4o-mini'
+		MAIN: 'claude-sonnet-4-20250514',
+		FALLBACK: 'claude-3-7-sonnet-20250219'
 	},
 	TASKS: {
 		DEFAULT_PRIORITY: 'medium' as TaskPriority,

scripts/modules/config-manager.js

@@ -27,21 +27,21 @@ const DEFAULTS = {
 	models: {
 		main: {
 			provider: 'anthropic',
-			modelId: 'claude-3-7-sonnet-20250219',
+			modelId: 'claude-sonnet-4-20250514',
 			maxTokens: 64000,
 			temperature: 0.2
 		},
 		research: {
 			provider: 'perplexity',
-			modelId: 'sonar-pro',
+			modelId: 'sonar',
 			maxTokens: 8700,
 			temperature: 0.1
 		},
 		fallback: {
 			// No default fallback provider/model initially
 			provider: 'anthropic',
-			modelId: 'claude-3-5-sonnet',
-			maxTokens: 8192, // Default parameters if fallback IS configured
+			modelId: 'claude-3-7-sonnet-20250219',
+			maxTokens: 120000, // Default parameters if fallback IS configured
 			temperature: 0.2
 		}
 	},

tests/unit/config-manager.test.js

@@ -119,20 +119,20 @@ const DEFAULT_CONFIG = {
 	models: {
 		main: {
 			provider: 'anthropic',
-			modelId: 'claude-3-7-sonnet-20250219',
+			modelId: 'claude-sonnet-4-20250514',
 			maxTokens: 64000,
 			temperature: 0.2
 		},
 		research: {
 			provider: 'perplexity',
-			modelId: 'sonar-pro',
+			modelId: 'sonar',
 			maxTokens: 8700,
 			temperature: 0.1
 		},
 		fallback: {
 			provider: 'anthropic',
-			modelId: 'claude-3-5-sonnet',
-			maxTokens: 8192,
+			modelId: 'claude-3-7-sonnet-20250219',
+			maxTokens: 120000,
 			temperature: 0.2
 		}
 	},