chore: add PRDs and surgical test generator agent

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Ralph Khreish <35776126+Crunchyman-ralph@users.noreply.github.com>

.changeset/chore-fix-docs.md

@@ -0,0 +1,5 @@
+---
+"task-master-ai": patch
+---
+
+Improve `analyze-complexity` cli docs and `--research` flag documentation

.changeset/config.json

@@ -11,9 +11,6 @@
   "access": "public",
   "baseBranch": "main",
   "ignore": [
-    "docs",
-    "@tm/cli",
-    "@tm/core",
-    "@tm/build-config"
+    "docs"
   ]
 }

.changeset/cursor-slash-commands.md

@@ -0,0 +1,7 @@
+---
+"task-master-ai": minor
+---
+
+Add Cursor IDE custom slash command support
+
+Expose Task Master commands as Cursor slash commands by copying assets/claude/commands to .cursor/commands on profile add and cleaning up on remove.

.changeset/curvy-weeks-flow.md

@@ -0,0 +1,5 @@
+---
+"task-master-ai": patch
+---
+
+Change parent task back to "pending" when all subtasks are in "pending" state

.changeset/easy-deer-heal.md

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

.changeset/mcp-timeout-configuration.md

@@ -0,0 +1,13 @@
+---
+"task-master-ai": minor
+---
+
+Enhanced Roo Code profile with MCP timeout configuration for improved reliability during long-running AI operations. The Roo profile now automatically configures a 300-second timeout for MCP server operations, preventing timeouts during complex tasks like `parse-prd`, `expand-all`, `analyze-complexity`, and `research` operations. This change also replaces static MCP configuration files with programmatic generation for better maintainability.
+
+**What's New:**
+- 300-second timeout for MCP operations (up from default 60 seconds)
+- Programmatic MCP configuration generation (replaces static asset files)
+- Enhanced reliability for AI-powered operations
+- Consistent with other AI coding assistant profiles
+
+**Migration:** No user action required - existing Roo Code installations will automatically receive the enhanced MCP configuration on next initialization.

.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/petite-ideas-grab.md

@@ -0,0 +1,5 @@
+---
+"task-master-ai": patch
+---
+
+Fix Claude Code settings validation for pathToClaudeCodeExecutable

.changeset/pre.json

@@ -1,17 +0,0 @@
-{
-  "mode": "pre",
-  "tag": "rc",
-  "initialVersions": {
-    "task-master-ai": "0.26.0",
-    "@tm/cli": "0.26.0",
-    "docs": "0.0.2",
-    "extension": "0.24.2",
-    "@tm/build-config": "1.0.0",
-    "@tm/core": "0.26.0"
-  },
-  "changesets": [
-    "moody-oranges-slide",
-    "odd-otters-tan",
-    "wild-ears-look"
-  ]
-}

.changeset/silly-pandas-find.md

@@ -0,0 +1,5 @@
+---
+"task-master-ai": patch
+---
+
+Fix sonar deep research model failing, should be called `sonar-deep-research`

.changeset/wild-ears-look.md

@@ -1,5 +0,0 @@
----
-"extension": minor
----
-
-Added a Start Build button to the VSCODE Task Properties Right Panel

.claude/agents/surgical-test-generator.md

@@ -0,0 +1,245 @@
+---
+name: surgical-test-generator
+description: Use this agent when you need to generate comprehensive, production-grade test suites for existing code. This includes: analyzing code paths and failure modes, detecting untestable patterns, generating unit/integration tests, following project test conventions, focusing on real-world edge cases, and ensuring deterministic test execution. The agent specializes in catching bugs that matter, not inflating coverage metrics.
+
+Examples:
+<example>
+Context: The user wants tests for a specific function they just wrote.
+user: "I need tests for this validatePayment function that handles credit card processing"
+assistant: "I'll use the surgical-test-generator agent to analyze the validatePayment function's code paths, identify critical edge cases like invalid card numbers and network failures, and generate comprehensive tests following your project's conventions."
+<commentary>
+User needs specific function testing - the agent will analyze paths, boundaries, and failure modes for payment validation.
+</commentary>
+</example>
+<example>
+Context: The user has untested legacy code that needs coverage.
+user: "We have this 500-line OrderProcessor class with zero tests and it's causing production issues"
+assistant: "I'll invoke the surgical-test-generator agent to map all the code paths in OrderProcessor, prioritize tests for the riskiest behaviors, and generate a test suite that catches the bugs most likely to cause production failures."
+<commentary>
+Legacy code without tests - the agent will prioritize high-risk paths and realistic failure scenarios over complete coverage.
+</commentary>
+</example>
+<example>
+Context: The user wants to improve test quality, not just coverage.
+user: "Our tests pass but we still get production bugs. Can you review this API handler and suggest better tests?"
+assistant: "I'll use the surgical-test-generator agent to identify gaps in your current test coverage, focusing on real-world edge cases like concurrent requests, malformed input, and external service failures that often slip through basic tests."
+<commentary>
+Existing tests missing edge cases - the agent will focus on adversarial scenarios and production-like failures.
+</commentary>
+</example>
+model: opus
+color: blue
+---
+
+You are a principal software engineer specializing in surgical, high-signal test generation. You write tests that catch real bugs before they reach production, focusing on actual failure modes over coverage metrics. You reason about control flow, data flow, mutation, concurrency, and security to design tests that surface defects early.
+
+## Core Capabilities
+
+### Multi-Perspective Analysis
+You sequentially analyze code through five expert lenses:
+1. **Context Profiling**: Identify language, frameworks, build tools, and existing test patterns
+2. **Path Analysis**: Map all reachable paths including happy, error, and exceptional flows
+3. **Adversarial Thinking**: Enumerate realistic failures, boundaries, and misuse patterns
+4. **Risk Prioritization**: Rank by production impact and likelihood, discard speculative cases
+5. **Test Scaffolding**: Generate deterministic, isolated tests following project conventions
+
+### Edge Case Expertise
+You focus on failures that actually occur in production:
+- **Data Issues**: Null/undefined, empty collections, malformed UTF-8, mixed line endings
+- **Numeric Boundaries**: -1, 0, 1, MAX values, floating-point precision, integer overflow
+- **Temporal Pitfalls**: DST transitions, leap years, timezone bugs, date parsing ambiguities
+- **Collection Problems**: Off-by-one errors, concurrent modification, performance at scale
+- **State Violations**: Out-of-order calls, missing initialization, partial updates
+- **External Failures**: Network timeouts, malformed responses, retry storms, connection exhaustion
+- **Concurrency Bugs**: Race conditions, deadlocks, promise leaks, thread starvation
+- **Resource Limits**: Memory spikes, file descriptor leaks, connection pool saturation
+- **Security Surfaces**: Injection attacks, path traversal, privilege escalation
+
+### Framework Intelligence
+You auto-detect and follow existing test patterns:
+- **JavaScript/TypeScript**: Jest, Vitest, Mocha, or project wrappers
+- **Python**: pytest, unittest with appropriate fixtures
+- **Java/Kotlin**: JUnit 5, TestNG with proper annotations
+- **C#/.NET**: xUnit.net preferred, NUnit/MSTest if dominant
+- **Go**: Built-in testing package with table-driven tests
+- **Rust**: Standard #[test] with proptest for properties
+- **Swift**: XCTest or Swift Testing based on project
+- **C/C++**: GoogleTest or Catch2 matching build system
+
+## Your Workflow
+
+### Phase 1: Code Analysis
+You examine the provided code to understand:
+- Public API surfaces and contracts
+- Internal helper functions and their criticality
+- External dependencies and I/O operations
+- State management and mutation patterns
+- Error handling and recovery paths
+- Concurrency and async operations
+
+### Phase 2: Test Strategy Development
+You determine the optimal testing approach:
+- Start from public APIs, work inward to critical helpers
+- Test behavior not implementation unless white-box needed
+- Prefer property-based tests for algebraic domains
+- Use minimal stubs/mocks, prefer in-memory fakes
+- Flag untestable code with refactoring suggestions
+- Include stress tests for concurrency issues
+
+### Phase 3: Test Generation
+You create tests that:
+- Follow project's exact style and conventions
+- Use clear Arrange-Act-Assert or Given-When-Then
+- Execute in under 100ms without external calls
+- Remain deterministic with seeded randomness
+- Self-document through descriptive names
+- Explain failures clearly with context
+
+## Detection Patterns
+
+### When analyzing a pure function:
+- Test boundary values and type edges
+- Verify mathematical properties hold
+- Check error propagation
+- Consider numeric overflow/underflow
+
+### When analyzing stateful code:
+- Test initialization sequences
+- Verify state transitions
+- Check concurrent access patterns
+- Validate cleanup and teardown
+
+### When analyzing I/O operations:
+- Test success paths with valid data
+- Simulate network failures and timeouts
+- Check malformed input handling
+- Verify resource cleanup on errors
+
+### When analyzing async code:
+- Test promise resolution and rejection
+- Check cancellation handling
+- Verify timeout behavior
+- Validate error propagation chains
+
+## Test Quality Standards
+
+### Coverage Philosophy
+You prioritize catching real bugs over metrics:
+- Critical paths get comprehensive coverage
+- Edge cases get targeted scenarios
+- Happy paths get basic validation
+- Speculative cases get skipped
+
+### Test Independence
+Each test you generate:
+- Runs in isolation without shared state
+- Cleans up all resources
+- Uses fixed seeds for randomness
+- Mocks time when necessary
+
+### Failure Diagnostics
+Your tests provide clear failure information:
+- Descriptive test names that explain intent
+- Assertions that show expected vs actual
+- Context about what was being tested
+- Hints about likely failure causes
+
+## Special Considerations
+
+### When NOT to Generate Tests
+You recognize when testing isn't valuable:
+- Generated code that's guaranteed correct
+- Simple getters/setters without logic
+- Framework boilerplate
+- Code scheduled for deletion
+
+### Untestable Code Patterns
+You identify and flag:
+- Hard-coded external dependencies
+- Global state mutations
+- Time-dependent behavior without injection
+- Random behavior without seeds
+
+### Performance Testing
+When relevant, you include:
+- Benchmarks for critical paths
+- Memory usage validation
+- Concurrent load testing
+- Resource leak detection
+
+## Output Format
+
+You generate test code that:
+
+```[language]
+// Clear test description
+test('should [specific behavior] when [condition]', () => {
+    // Arrange - Set up test data and dependencies
+    
+    // Act - Execute the code under test
+    
+    // Assert - Verify the outcome
+});
+```
+
+## Framework-Specific Patterns
+
+### JavaScript/TypeScript (Jest/Vitest)
+- Use describe blocks for grouping
+- Leverage beforeEach/afterEach for setup
+- Mock modules with jest.mock()/vi.fn, vi.mock or vi.spyOn
+- Use expect matchers appropriately
+
+### Python (pytest)
+- Use fixtures for reusable setup
+- Apply parametrize for data-driven tests
+- Leverage pytest markers for categorization
+- Use clear assertion messages
+
+### Java (JUnit 5)
+- Apply appropriate annotations
+- Use nested classes for grouping
+- Leverage parameterized tests
+- Include display names
+
+### C# (xUnit)
+- Use Theory for data-driven tests
+- Apply traits for categorization
+- Leverage fixtures for setup
+- Use fluent assertions when available
+
+## Request Handling
+
+### Specific Test Requests
+When asked for specific tests:
+- Focus only on the requested scope
+- Don't generate broader coverage unless asked
+- Provide targeted, high-value scenarios
+- Include rationale for test choices
+
+### Comprehensive Coverage Requests
+When asked for full coverage:
+- Start with critical paths
+- Add edge cases progressively
+- Group related tests logically
+- Flag if multiple files needed
+
+### Legacy Code Testing
+When testing untested code:
+- Prioritize high-risk areas first
+- Add characterization tests
+- Suggest refactoring for testability
+- Focus on preventing regressions
+
+## Communication Guidelines
+
+You always:
+- Explain why each test matters
+- Document test intent clearly
+- Note any assumptions made
+- Highlight untestable patterns
+- Suggest improvements when relevant
+- Follow existing project style exactly
+- Generate only essential test code
+
+When you need additional context like test frameworks or existing patterns, you ask specifically for those files. You focus on generating tests that will actually catch bugs in production, not tests that merely increase coverage numbers. Every test you write has a clear purpose and tests a realistic scenario.

.cursor/mcp.json

@@ -2,7 +2,7 @@
 	"mcpServers": {
 		"task-master-ai": {
 			"command": "node",
-			"args": ["./mcp-server/server.js"],
+			"args": ["./dist/mcp-server.js"],
 			"env": {
 				"ANTHROPIC_API_KEY": "ANTHROPIC_API_KEY_HERE",
 				"PERPLEXITY_API_KEY": "PERPLEXITY_API_KEY_HERE",

.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/ci.yml

@@ -6,9 +6,6 @@ on:
       - main
       - next
   pull_request:
-    branches:
-      - main
-      - next
   workflow_dispatch:
 
 concurrency:
@@ -92,6 +89,9 @@ jobs:
         env:
           NODE_ENV: production
           FORCE_COLOR: 1
+          TM_PUBLIC_BASE_DOMAIN: ${{ secrets.TM_PUBLIC_BASE_DOMAIN }}
+          TM_PUBLIC_SUPABASE_URL: ${{ secrets.TM_PUBLIC_SUPABASE_URL }}
+          TM_PUBLIC_SUPABASE_ANON_KEY: ${{ secrets.TM_PUBLIC_SUPABASE_ANON_KEY }}
 
       - name: Upload build artifacts
         uses: actions/upload-artifact@v4

.github/workflows/pre-release.yml

@@ -75,6 +75,9 @@ jobs:
         env:
           NODE_ENV: production
           FORCE_COLOR: 1
+          TM_PUBLIC_BASE_DOMAIN: ${{ secrets.TM_PUBLIC_BASE_DOMAIN }}
+          TM_PUBLIC_SUPABASE_URL: ${{ secrets.TM_PUBLIC_SUPABASE_URL }}
+          TM_PUBLIC_SUPABASE_ANON_KEY: ${{ secrets.TM_PUBLIC_SUPABASE_ANON_KEY }}
 
       - name: Create Release Candidate Pull Request or Publish Release Candidate to npm
         uses: changesets/action@v1

.github/workflows/release.yml

@@ -22,7 +22,7 @@ jobs:
       - uses: actions/setup-node@v4
         with:
           node-version: 20
-          cache: 'npm'
+          cache: "npm"
 
       - name: Cache node_modules
         uses: actions/cache@v4
@@ -46,6 +46,9 @@ jobs:
         env:
           NODE_ENV: production
           FORCE_COLOR: 1
+          TM_PUBLIC_BASE_DOMAIN: ${{ secrets.TM_PUBLIC_BASE_DOMAIN }}
+          TM_PUBLIC_SUPABASE_URL: ${{ secrets.TM_PUBLIC_SUPABASE_URL }}
+          TM_PUBLIC_SUPABASE_ANON_KEY: ${{ secrets.TM_PUBLIC_SUPABASE_ANON_KEY }}
 
       - name: Create Release Pull Request or Publish to npm
         uses: changesets/action@v1

.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 }}
+            • Avg Response Time: ${{ steps.metrics.outputs.pr_avg_first_response }}
+            • Avg Time to Merge: ${{ steps.metrics.outputs.pr_avg_merge_time }}
 
-            **⏱️ 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 }}
+            **📈 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

.kiro/settings/mcp.json

@@ -2,7 +2,7 @@
 	"mcpServers": {
 		"task-master-ai": {
 			"command": "npx",
-			"args": ["-y", "--package=task-master-ai", "task-master-ai"],
+			"args": ["-y", "task-master-ai"],
 			"env": {
 				"ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
 				"PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY_HERE",

.manypkg.json

@@ -0,0 +1,6 @@
+{
+	"$schema": "https://unpkg.com/@manypkg/get-packages@1.1.3/schema.json",
+	"defaultBranch": "main",
+	"ignoredRules": ["ROOT_HAS_DEPENDENCIES", "INTERNAL_MISMATCH"],
+	"ignoredPackages": ["@tm/core", "@tm/cli", "@tm/build-config"]
+}

.taskmaster/CLAUDE.md

@@ -85,7 +85,7 @@ Task Master provides an MCP server that Claude Code can connect to. Configure in
   "mcpServers": {
     "task-master-ai": {
       "command": "npx",
-      "args": ["-y", "--package=task-master-ai", "task-master-ai"],
+      "args": ["-y", "task-master-ai"],
       "env": {
         "ANTHROPIC_API_KEY": "your_key_here",
         "PERPLEXITY_API_KEY": "your_key_here",

.taskmaster/config.json

@@ -2,8 +2,8 @@
 	"models": {
 		"main": {
 			"provider": "anthropic",
-			"modelId": "claude-3-7-sonnet-20250219",
-			"maxTokens": 120000,
+			"modelId": "claude-sonnet-4-20250514",
+			"maxTokens": 64000,
 			"temperature": 0.2
 		},
 		"research": {
@@ -14,8 +14,8 @@
 		},
 		"fallback": {
 			"provider": "anthropic",
-			"modelId": "claude-3-5-sonnet-20241022",
-			"maxTokens": 8192,
+			"modelId": "claude-3-7-sonnet-20250219",
+			"maxTokens": 120000,
 			"temperature": 0.2
 		}
 	},
@@ -29,9 +29,15 @@
 		"ollamaBaseURL": "http://localhost:11434/api",
 		"bedrockBaseURL": "https://bedrock.us-east-1.amazonaws.com",
 		"responseLanguage": "English",
+		"enableCodebaseAnalysis": true,
 		"userId": "1234567890",
 		"azureBaseURL": "https://your-endpoint.azure.com/",
 		"defaultTag": "master"
 	},
-	"claudeCode": {}
+	"claudeCode": {},
+	"grokCli": {
+		"timeout": 120000,
+		"workingDirectory": null,
+		"defaultModel": "grok-4-latest"
+	}
 }

.taskmaster/docs/prd-autonomous-tdd-rails.md

@@ -0,0 +1,220 @@
+# PRD: Autonomous TDD + Git Workflow (On Rails)
+
+## Summary
+- Put the existing git and test workflows on rails: a repeatable, automated process that can run autonomously, with guardrails and a compact TUI for visibility.
+- Flow: for a selected task, create a branch named with the tag + task id → generate tests for the first subtask (red) using the Surgical Test Generator → implement code (green) → verify tests → commit → repeat per subtask → final verify → push → open PR against the default branch.
+- Build on existing rules: `.cursor/rules/git_workflow.mdc`, `.cursor/rules/test_workflow.mdc`, `.claude/agents/surgical-test-generator.md`, and existing CLI/core services.
+
+## Goals
+- Deterministic, resumable automation to execute the TDD loop per subtask with minimal human intervention.
+- Strong guardrails: never commit to the default branch; only commit when tests pass; enforce status transitions; persist logs/state for debuggability.
+- Visibility: a compact terminal UI (like lazygit) to pick tag, view tasks, and start work; right-side pane opens an executor terminal (via tmux) for agent coding.
+- Extensible: framework-agnostic test generation via the Surgical Test Generator; detect and use the repo’s test command for execution with coverage thresholds.
+
+## Non‑Goals (initial)
+- Full multi-language runner parity beyond detection and executing the project’s test command.
+- Complex GUI; start with CLI/TUI + tmux pane. IDE/extension can hook into the same state later.
+- Rich executor selection UX (codex/gemini/claude) — we’ll prompt per run; defaults can come later.
+
+## Success Criteria
+- One command can autonomously complete a task’s subtasks via TDD and open a PR when done.
+- All commits made on a branch that includes the tag and task id (see Branch Naming); no commits to the default branch directly.
+- Every subtask iteration: failing tests added first (red), then code added to pass them (green), commit only after green.
+- End-to-end logs + artifacts stored in `.taskmaster/reports/runs/<timestamp-or-id>/`.
+
+## User Stories
+- As a developer, I can run `tm autopilot <taskId>` and watch a structured, safe workflow execute.
+- As a reviewer, I can inspect commits per subtask, and a PR summarizing the work when the task completes.
+- As an operator, I can see current step, active subtask, tests status, and logs in a compact CLI view and read a final run report.
+
+## High‑Level Workflow
+1) Pre‑flight
+   - Verify clean working tree or confirm staging/commit policy (configurable).
+   - Detect repo type and the project’s test command (e.g., `npm test`, `pnpm test`, `pytest`, `go test`).
+   - Validate tools: `git`, `gh` (optional for PR), `node/npm`, and (if used) `claude` CLI.
+   - Load TaskMaster state and selected task; if no subtasks exist, automatically run “expand” before working.
+
+2) Branch & Tag Setup
+   - Checkout default branch and update (optional), then create a branch using Branch Naming (below).
+   - Map branch ↔ tag via existing tag management; explicitly set active tag to the branch’s tag.
+
+3) Subtask Loop (for each pending/in-progress subtask in dependency order)
+   - Select next eligible subtask using `tm-core` TaskService `getNextTask()` and subtask eligibility logic.
+   - Red: generate or update failing tests for the subtask
+     - Use the Surgical Test Generator system prompt (`.claude/agents/surgical-test-generator.md`) to produce high-signal tests following project conventions.
+     - Run tests to confirm red; record results. If not red (already passing), skip to next subtask or escalate.
+   - Green: implement code to pass tests
+     - Use executor to implement changes (initial: `claude` CLI prompt with focused context).
+     - Re-run tests until green or timeout/backoff policy triggers.
+   - Commit: when green
+     - Commit tests + code with conventional commit message. Optionally update subtask status to `done`.
+     - Persist run step metadata/logs.
+
+4) Finalization
+   - Run full test suite and coverage (if configured); optionally lint/format.
+   - Commit any final adjustments.
+   - Push branch (ask user to confirm); create PR (via `gh pr create`) targeting the default branch. Title format: `Task #<id> [<tag>]: <title>`.
+
+5) Post‑Run
+   - Update task status if desired (e.g., `review`).
+   - Persist run report (JSON + markdown summary) to `.taskmaster/reports/runs/<run-id>/`.
+
+## Guardrails
+- Never commit to the default branch.
+- Commit only if all tests (targeted and suite) pass; allow override flags.
+- Enforce 80% coverage thresholds (lines/branches/functions/statements) by default; configurable.
+- Timebox/model ops and retries; if not green within N attempts, pause with actionable state for resume.
+- Always log actions, commands, and outcomes; include dry-run mode.
+- Ask before branch creation, pushing, and opening a PR unless `--no-confirm` is set.
+
+## Integration Points (Current Repo)
+- CLI: `apps/cli` provides command structure and UI components.
+  - New command: `tm autopilot` (alias: `task-master autopilot`).
+  - Reuse UI components under `apps/cli/src/ui/components/` for headers/task details/next-task.
+- Core services: `packages/tm-core`
+  - `TaskService` for selection, status, tags.
+  - `TaskExecutionService` for prompt formatting and executor prep.
+  - Executors: `claude` executor and `ExecutorFactory` to run external tools.
+  - Proposed new: `WorkflowOrchestrator` to drive the autonomous loop and emit progress events.
+- Tag/Git utilities: `scripts/modules/utils/git-utils.js` and `scripts/modules/task-manager/tag-management.js` for branch→tag mapping and explicit tag switching.
+- Rules: `.cursor/rules/git_workflow.mdc` and `.cursor/rules/test_workflow.mdc` to steer behavior and ensure consistency.
+- Test generation prompt: `.claude/agents/surgical-test-generator.md`.
+
+## Proposed Components
+- Orchestrator (tm-core): `WorkflowOrchestrator` (new)
+  - State machine driving phases: Preflight → Branch/Tag → SubtaskIter (Red/Green/Commit) → Finalize → PR.
+  - Exposes an evented API (progress events) that the CLI can render.
+  - Stores run state artifacts.
+
+- Test Runner Adapter
+  - Detects and runs tests via the project’s test command (e.g., `npm test`), with targeted runs where feasible.
+  - API: runTargeted(files/pattern), runAll(), report summary (failures, duration, coverage), enforce 80% threshold by default.
+
+- Git/PR Adapter
+  - Encapsulates `git` ops: branch create/checkout, add/commit, push.
+  - Optional `gh` integration to open PR; fallback to instructions if `gh` unavailable.
+  - Confirmation gates for branch creation and pushes.
+  - Adds commit footers and a unified trailer (`Refs: TM-<tag>-<id>[.<sub>]`) for robust mapping to tasks/subtasks.
+
+- Prompt/Exec Adapter
+  - Uses existing executor service to call the selected coding assistant (initially `claude`) with tight prompts: task/subtask context, surgical tests first, then minimal code to green.
+
+- Run State + Reporting
+  - JSONL log of steps, timestamps, commands, test results.
+  - Markdown summary for PR description and post-run artifact.
+
+## CLI UX (MVP)
+- Command: `tm autopilot [taskId]`
+  - Flags: `--dry-run`, `--no-push`, `--no-pr`, `--no-confirm`, `--force`, `--max-attempts <n>`, `--runner <auto|custom>`, `--commit-scope <scope>`
+  - Output: compact header (project, tag, branch), current phase, subtask line, last test summary, next actions.
+- Resume: If interrupted, `tm autopilot --resume` picks up from last checkpoint in run state.
+
+### TUI with tmux (Linear Execution)
+- Left pane: Tag selector, task list (status/priority), start/expand shortcuts; “Start” triggers the next task or a selected task.
+- Right pane: Executor terminal (tmux split) that runs the coding agent (claude-code/codex). Autopilot can hand over to the right pane during green.
+- MCP integration: use MCP tools for task queries/updates and for shell/test invocations where available.
+
+## Prompts (Initial Direction)
+- Red phase prompt skeleton (tests):
+  - Use `.claude/agents/surgical-test-generator.md` as the system prompt to generate high-signal failing tests tailored to the project’s language and conventions. Keep scope minimal and deterministic; no code changes yet.
+- Green phase prompt skeleton (code):
+  - “Make the tests pass by changing the smallest amount of code, following project patterns. Only modify necessary files. Keep commits focused to this subtask.”
+
+## Configuration
+- `.taskmaster/config.json` additions
+  - `autopilot`: `{ enabled: true, requireCleanWorkingTree: true, commitTemplate: "{type}({scope}): {msg}", defaultCommitType: "feat" }`
+  - `test`: `{ runner: "auto", coverageThresholds: { lines: 80, branches: 80, functions: 80, statements: 80 } }`
+  - `git`: `{ branchPattern: "{tag}/task-{id}-{slug}", pr: { enabled: true, base: "default" }, commitFooters: { task: "Task-Id", subtask: "Subtask-Id", tag: "Tag" }, commitTrailer: "Refs: TM-{tag}-{id}{.sub?}" }`
+
+## Decisions
+- Use conventional commits plus footers and a unified trailer `Refs: TM-<tag>-<id>[.<sub>]` for all commits; Git/PR adapter is responsible for injecting these.
+
+## Risks and Mitigations
+- Model hallucination/large diffs: restrict prompt scope; enforce minimal changes; show diff previews (optional) before commit.
+- Flaky tests: allow retries, isolate targeted runs for speed, then full suite before commit.
+- Environment variability: detect runners/tools; provide fallbacks and actionable errors.
+- PR creation fails: still push and print manual commands; persist PR body to reuse.
+
+## Open Questions
+1) Slugging rules for branch names; any length limits or normalization beyond `{slug}` token sanitize?
+2) PR body standard sections beyond run report (e.g., checklist, coverage table)?
+3) Default executor prompt fine-tuning once codex/gemini integration is available.
+4) Where to store persistent TUI state (pane layout, last selection) in `.taskmaster/state.json`?
+
+## Branch Naming
+- Include both the tag and the task id in the branch name to make lineage explicit.
+- Default pattern: `<tag>/task-<id>[-slug]` (e.g., `master/task-12`, `tag-analytics/task-4-user-auth`).
+- Configurable via `.taskmaster/config.json`: `git.branchPattern` supports tokens `{tag}`, `{id}`, `{slug}`.
+
+## PR Base Branch
+- Use the repository’s default branch (detected via git) unless overridden.
+- Title format: `Task #<id> [<tag>]: <title>`.
+
+## RPG Mapping (Repository Planning Graph)
+
+Functional nodes (capabilities):
+- Autopilot Orchestration → drives TDD loop and lifecycle
+- Test Generation (Surgical) → produces failing tests from subtask context
+- Test Execution + Coverage → runs suite, enforces thresholds
+- Git/Branch/PR Management → safe operations and PR creation
+- TUI/Terminal Integration → interactive control and visibility via tmux
+- MCP Integration → structured task/status/context operations
+
+Structural nodes (code organization):
+- `packages/tm-core`:
+  - `services/workflow-orchestrator.ts` (new)
+  - `services/test-runner-adapter.ts` (new)
+  - `services/git-adapter.ts` (new)
+  - existing: `task-service.ts`, `task-execution-service.ts`, `executors/*`
+- `apps/cli`:
+  - `src/commands/autopilot.command.ts` (new)
+  - `src/ui/tui/` (new tmux/TUI helpers)
+- `scripts/modules`:
+  - reuse `utils/git-utils.js`, `task-manager/tag-management.js`
+- `.claude/agents/`:
+  - `surgical-test-generator.md`
+
+Edges (data/control flow):
+- Autopilot → Test Generation → Test Execution → Git Commit → loop
+- Autopilot → Git Adapter (branch, tag, PR)
+- Autopilot → TUI (event stream) → tmux pane control
+- Autopilot → MCP tools for task/status updates
+- Test Execution → Coverage gate → Autopilot decision
+
+Topological traversal (implementation order):
+1) Git/Test adapters (foundations)
+2) Orchestrator skeleton + events
+3) CLI `autopilot` command and dry-run
+4) Surgical test-gen integration and execution gate
+5) PR creation, run reports, resumability
+
+## Phased Roadmap
+- Phase 0: Spike
+  - Implement CLI skeleton `tm autopilot` with dry-run showing planned steps from a real task + subtasks.
+  - Detect test runner (package.json) and git state; render a preflight report.
+
+- Phase 1: Core Rails
+  - Implement `WorkflowOrchestrator` in `tm-core` with event stream; add Git/Test adapters.
+  - Support subtask loop (red/green/commit) with framework-agnostic test generation and detected test command; commit gating on passing tests and coverage.
+  - Branch/tag mapping via existing tag-management APIs.
+  - Run report persisted under `.taskmaster/reports/runs/`.
+
+- Phase 2: PR + Resumability
+  - Add `gh` PR creation with well-formed body using the run report.
+  - Introduce resumable checkpoints and `--resume` flag.
+  - Add coverage enforcement and optional lint/format step.
+
+- Phase 3: Extensibility + Guardrails
+  - Add support for basic pytest/go test adapters.
+  - Add safeguards: diff preview mode, manual confirm gates, aggressive minimal-change prompts.
+  - Optional: small TUI panel and extension panel leveraging the same run state file.
+
+## References (Repo)
+- Test Workflow: `.cursor/rules/test_workflow.mdc`
+- Git Workflow: `.cursor/rules/git_workflow.mdc`
+- CLI: `apps/cli/src/commands/start.command.ts`, `apps/cli/src/ui/components/*.ts`
+- Core Services: `packages/tm-core/src/services/task-service.ts`, `task-execution-service.ts`
+- Executors: `packages/tm-core/src/executors/*`
+- Git Utilities: `scripts/modules/utils/git-utils.js`
+- Tag Management: `scripts/modules/task-manager/tag-management.js`
+ - Surgical Test Generator: `.claude/agents/surgical-test-generator.md`

.taskmaster/docs/prd-rpg-user-stories.md

@@ -0,0 +1,221 @@
+# PRD: RPG‑Based User Story Mode + Validation‑First Delivery
+
+## Summary
+- Introduce a “User Story Mode” where each Task is a user story and each Subtask is a concrete implementation step. Enable via config flag; when enabled, Task generation and PRD parsing produce user‑story titles/details with acceptance criteria, while Subtasks capture implementation details.
+- Build a validation‑first delivery pipeline: derive tests from acceptance criteria (Surgical Test Generator), wire TDD rails and Git/PR mapping so reviews focus on verification rather than code spelunking.
+- Keep everything on rails: branch naming with tag+task id, commit/PR linkage to tasks/subtasks, coverage + test gates, and lightweight TUI for fast execution.
+
+## North‑Star Outcomes
+- Humans stay in briefs/frontends; implementation runs quickly, often without opening the IDE.
+- “Definition of Done” is expressed and enforced as tests; business logic is encoded in test criteria/acceptance criteria.
+- End‑to‑end linkage from brief → user story → subtasks → commits/PRs → delivery, with reproducible automation and minimal ceremony.
+
+## Problem
+- The bottleneck is validation and PR review, not code generation. Plans are helpful but the chokepoint is proving correctness, business conformance, and integration.
+- Current test workflow is too Jest‑specific; we need framework‑agnostic generation and execution.
+- We need consistent Git/TDD wiring so GitHub integrations can map work artifacts to tasks/subtasks without ambiguity.
+
+## Solution Overview
+- Add a configuration flag to switch to user story mode and adapt prompts/parsers.
+- Expand tasks with explicit Acceptance Criteria and Test Criteria; drive Surgical Test Generator to create failing tests first; wire autonomous TDD loops per subtask until green, then commit.
+- Enforce coverage (80% default) and generate PRs that summarize user story, acceptance criteria coverage, and test results; commits/PRs contain metadata to link back to tasks/subtasks.
+- Provide a compact TUI (tmux) to pick tag/tasks and launch an executor terminal, while the orchestrator runs rails in the background.
+
+---
+
+## Configuration
+- `.taskmaster/config.json` additions
+  - `stories`: `{ enabled: true, storyLabel: "User Story", acceptanceKey: "Acceptance Criteria" }`
+  - `autopilot`: `{ enabled: true, requireCleanWorkingTree: true }`
+  - `test`: `{ runner: "auto", coverageThresholds: { lines: 80, branches: 80, functions: 80, statements: 80 } }`
+  - `git`: `{ branchPattern: "{tag}/task-{id}-{slug}", pr: { enabled: true, base: "default" }, commitFooters: { task: "Task-Id", subtask: "Subtask-Id", tag: "Tag" } }`
+
+Behavior when `stories.enabled=true`:
+- Task generation prompts and PRD parsers produce user‑story formatted titles and descriptions, include acceptance criteria blocks, and set `task.type = 'user_story'`.
+- Subtasks remain implementation steps with concise technical goals.
+- Expand will ensure any missing acceptance criteria is synthesized (from brief/PRD content) before starting work.
+
+---
+
+## Data Model Changes
+- Task fields (add):
+  - `type: 'user_story' | 'technical'` (default `technical`)
+  - `acceptanceCriteria: string[] | string` (structured or markdown)
+  - `testCriteria: string[] | string` (optional, derived from acceptance criteria; what to validate)
+- Subtask fields remain focused on implementation detail and dependency graph.
+
+Storage and UI remain backward compatible; fields are optional when `stories.enabled=false`.
+
+### JSON Gherkin Representation (for stories)
+Add an optional `gherkin` block to Tasks in story mode. Keep Hybrid acceptanceCriteria as the human/authoring surface; maintain a normalized JSON Gherkin for deterministic mapping.
+
+```
+GherkinFeature {
+  id: string,                   // FEAT-<taskId>
+  name: string,                 // mirrors user story title
+  description?: string,
+  background?: { steps: Step[] },
+  scenarios: Scenario[]
+}
+
+Scenario {
+  id: string,                   // SC-<taskId>-<n> or derived from AC id
+  name: string,
+  tags?: string[],
+  steps: Step[],                // Given/When/Then/And/But
+  examples?: Record<string, any>[]
+}
+
+Step { keyword: 'Given'|'When'|'Then'|'And'|'But', text: string }
+```
+
+Notes
+- Derive `gherkin.scenarios` from acceptanceCriteria when obvious; preserve both raw markdown and normalized items.
+- Allow cross‑references between scenarios and AC items (e.g., `refs: ['AC-12-1']`).
+
+---
+
+## RPG Plan (Repository Planning Graph)
+
+Functional Nodes (Capabilities)
+- Brief Intake → parse briefs/PRDs and extract user stories (when enabled)
+- User Story Generation → create task title/details as user stories + acceptance criteria
+- JSON Gherkin Synthesis → produce Feature/Scenario structure from acceptance criteria
+- Acceptance/Test Criteria Synthesis → convert acceptance criteria into concrete test criteria
+- Surgical Test Generation → generate failing tests per subtask using `.claude/agents/surgical-test-generator.md`
+- Implementation Planning → expand subtasks as atomic implementation steps with dependencies
+- Autonomous Execution (Rails) → branch, red/green loop per subtask, commit when green
+- Validation & Review Automation → coverage gates, PR body with user story + results, checklist
+- GitHub Integration Mapping → branch naming, commit footers, PR linkage to tasks/subtasks
+- TUI/Terminal Integration → tag/task selection left pane; executor terminal right pane via tmux
+
+Structural Nodes (Code Organization)
+- `packages/tm-core`
+  - `services/workflow-orchestrator.ts` (new): drives rails, emits progress events
+  - `services/story-mode.service.ts` (new): toggles prompts/parsers for user stories, acceptance criteria
+  - `services/test-runner-adapter.ts` (new): detects/executes project test command, collects coverage
+  - `services/git-adapter.ts` (new): branch/commit/push, PR creation; applies commit footers
+  - existing: `task-service.ts`, `task-execution-service.ts`, `executors/*`
+- `apps/cli`
+  - `src/commands/autopilot.command.ts` (new): orchestrates a full run; supports `--stories`
+  - `src/ui/tui/` (new): tmux helpers and compact panes for selection and logs
+- `scripts/modules`
+  - reuse `utils/git-utils.js`, `task-manager/tag-management.js`, PR template utilities
+- `.cursor/rules`
+  - update generation/parsing rules to emit user‑story format when enabled
+- `.claude/agents`
+  - existing: `surgical-test-generator.md` for red phase
+
+Edges (Dependencies / Data Flow)
+- Brief Intake → User Story Generation → Acceptance/Test Criteria Synthesis → Implementation Planning → Autonomous Execution → Validation/PR
+- Execution ↔ Test Runner (runAll/runTargeted, coverage) → back to Execution for decisions
+- Git Adapter ← Execution (commits/branch) → PR creation (target default branch)
+- TUI ↔ Orchestrator (event stream) → user confirmations for branch/push/PR
+- MCP Tools ↔ Orchestrator for task/status/context updates
+
+Topological Traversal (Build Order)
+1) Config + Data Model changes (stories flag, acceptance fields, optional `gherkin`)
+2) Rules/Prompts updates for parsing/generation in story mode (emit AC Hybrid + JSON Gherkin)
+3) Test Runner Adapter (framework‑agnostic execute + coverage)
+4) Git Adapter (branch pattern `{tag}/task-{id}-{slug}`, commit footers/trailer, PR create)
+5) Workflow Orchestrator wiring red/green/commit loop with coverage gate and scenario iteration
+6) Surgical Test Gen integration (red) from JSON Gherkin + AC; minimal‑change impl prompts (green)
+7) CLI autopilot (dry‑run → full run) and TUI (tmux panes)
+8) PR template and review automation (user story, AC table with test links, scenarios, coverage)
+
+---
+
+## Git/TDD Wiring (Validation‑First)
+- Branch naming: include tag + task id (e.g., `master/task-12-user-auth`) to disambiguate context.
+- Commit footers (configurable):
+  - `Task-Id: <id>`
+  - `Subtask-Id: <id>.<sub>` when relevant
+  - `Tag: <tag>`
+  - Trailer: `Refs: TM-<tag>-<id>[.<sub>] SC:<scenarioId> AC:<acId>`
+- Red/Green/Commit loop per subtask:
+  - Red: synthesize failing tests from acceptance criteria (Surgical agent)
+  - Green: minimal code to pass; re‑run full suite
+  - Commit when all tests pass and coverage ≥ 80%
+- PR base: repository default branch. Title `Task #<id> [<tag>]: <title>`.
+- PR body sections: User Story, Acceptance Criteria, Subtask Summary, Test Results, Coverage Table, Linked Work Items (ids), Next Steps.
+
+---
+
+## Prompts & Parsers (Story Mode)
+- PRD/Brief Parser updates:
+  - Extract user stories with “As a … I want … so that …” format when present.
+  - Extract Acceptance Criteria as bullet list; fill gaps with LLM synthesis from brief context.
+  - Emit JSON Gherkin Feature/Scenarios; auto‑split Given/When/Then when feasible; otherwise store text under `then` and refine later.
+- Task Generation Prompt (story mode):
+  - “Generate a task as a User Story with clear Acceptance Criteria. Do not include implementation details in the story; produce implementation subtasks separately.”
+- Subtask Generation Prompt:
+  - “Produce technical implementation steps to satisfy the acceptance criteria. Each subtask should be atomic and testable.”
+- Test Generation (Red):
+  - Use `.claude/agents/surgical-test-generator.md`; seed with JSON Gherkin + Acceptance/Test Criteria; determinism favored over maximum coverage.
+  - Record produced test paths back into AC items and optionally scenario annotations.
+- Implementation (Green):
+  - Minimal diffs, follow patterns, keep commits scoped to the subtask.
+
+---
+
+## TUI (Linear, tmux‑based)
+- Left: Tag selector and task list (status/priority). Actions: Expand, Start (Next or Selected), Review.
+- Right: Executor terminal (claude‑code/codex) under tmux split; orchestrator logs under another pane.
+- Confirmations inline (branch create, push, PR) unless `--no-confirm`.
+
+---
+
+## Migration & Backward Compatibility
+- Optional `gherkin` block; existing tasks remain valid.
+- When `stories.enabled=true`, new tasks include AC Hybrid + `gherkin`; upgrade path via a utility to synthesize both from description/testStrategy/acceptanceCriteria.
+
+---
+
+## Risks & Mitigations
+- Hallucinated acceptance criteria → Always show criteria in PR; allow quick amend and re‑run.
+- Framework variance → Test Runner Adapter detects and normalizes execution/coverage; fallback to `test` script.
+- Large diffs → Prompt for minimal changes; allow diff preview before commit.
+- Flaky tests → Retry policy; isolate targeted runs; enforce passing full suite before commit.
+
+---
+
+## Acceptance Criteria Schema Options (for decision)
+- Option A: Markdown only
+  - Pros: simple to write/edit, good for humans
+  - Cons: hard to map deterministically to tests; weak traceability; brittle diffs
+- Option B: Structured array
+  - Example: `{ id, summary, given, when, then, severity, tags }`
+  - Pros: machine‑readable; strong linking to tests/coverage; easy to diff
+  - Cons: heavier authoring; requires schema discipline
+- Option C: Hybrid (recommended)
+  - Store both: a normalized array of criteria objects and a preserved `raw` markdown block
+  - Each criterion gets a stable `id` (e.g., `AC-<taskId>-<n>`) used in tests, commit trailers, and PR tables
+  - Enables clean PR tables and deterministic coverage mapping while keeping human‑friendly text
+
+Proposed default schema (hybrid):
+```
+acceptanceCriteria: {
+  raw: """
+  - AC1: Guest can checkout with credit card
+  - AC2: Declined cards show error inline
+  """,
+  items: [
+    {
+      id: "AC-12-1",
+      summary: "Guest can checkout with credit card",
+      given: "a guest with items in cart",
+      when: "submits valid credit card",
+      then: "order is created and receipt emailed",
+      severity: "must",
+      tags: ["checkout", "payments"],
+      tests: [] // filled by orchestrator (file paths/test IDs)
+    }
+  ]
+}
+```
+
+Decision: adopt Hybrid default; allow Markdown‑only input and auto‑normalize.
+
+## Decisions
+- Adopt Hybrid acceptance criteria schema by default; normalize Markdown to structured items with stable IDs `AC-<taskId>-<n>`.
+- Use conventional commits plus footers and a unified trailer `Refs: TM-<tag>-<id>[.<sub>]` across PRDs for robust mapping.

.taskmaster/docs/prd-tm-start.txt

@@ -0,0 +1,91 @@
+<context>
+# Overview
+Add a new CLI command: `task-master start <task_id>` (alias: `tm start <task_id>`). This command hard-codes `claude-code` as the executor, fetches task details, builds a standardized prompt, runs claude-code, shows the result, checks for git changes, and auto-marks the task as done if successful.
+
+We follow the Commander class pattern, reuse task retrieval from `show` command flow. Extremely minimal for 1-hour hackathon timeline.
+
+# Core Features
+- `start` command (Commander class style) 
+- Hard-coded executor: `claude-code`
+- Standardized prompt designed for minimal changes following existing patterns
+- Shows claude-code output (no streaming)
+- Git status check for success detection
+- Auto-mark task done if successful
+
+# User Experience
+```
+task-master start 12
+```
+1) Fetches Task #12 details
+2) Builds standardized prompt with task context
+3) Runs claude-code with the prompt
+4) Shows output
+5) Checks git status for changes
+6) Auto-marks task done if changes detected
+</context>
+
+<PRD>
+# Technical Architecture
+
+- Command pattern:
+  - Create `apps/cli/src/commands/start.command.ts` modeled on [list.command.ts](mdc:apps/cli/src/commands/list.command.ts) and task lookup from [show.command.ts](mdc:apps/cli/src/commands/show.command.ts)
+
+- Task retrieval:
+  - Use `@tm/core` via `createTaskMasterCore` to get task by ID
+  - Extract: id, title, description, details
+
+- Executor (ultra-simple approach):
+  - Execute `claude "full prompt here"` command directly
+  - The prompt tells Claude to first run `tm show <task_id>` to get task details
+  - Then tells Claude to implement the code changes
+  - This opens Claude CLI interface naturally in the current terminal
+  - No subprocess management needed - just execute the command
+
+- Execution flow:
+  1) Validate `<task_id>` exists; exit with error if not
+  2) Build standardized prompt that includes instructions to run `tm show <task_id>`
+  3) Execute `claude "prompt"` command directly in terminal
+  4) Claude CLI opens, runs `tm show`, then implements changes
+  5) After Claude session ends, run `git status --porcelain` to detect changes
+  6) If changes detected, auto-run `task-master set-status --id=<task_id> --status=done`
+
+- Success criteria:
+  - Success = exit code 0 AND git shows modified/created files
+  - Print changed file paths; warn if no changes detected
+
+# Development Roadmap
+
+MVP (ship in ~1 hour):
+1) Implement `start.command.ts` (Commander class), parse `<task_id>`
+2) Validate task exists via tm-core 
+3) Build prompt that tells Claude to run `tm show <task_id>` then implement
+4) Execute `claude "prompt"` command, then check git status and auto-mark done
+
+# Risks and Mitigations
+- Executor availability: Error clearly if `claude-code` provider fails
+- False success: Git-change heuristic acceptable for hackathon MVP
+
+# Appendix
+
+**Standardized Prompt Template:**
+```
+You are an AI coding assistant with access to this repository's codebase.
+
+First, run this command to get the task details:
+tm show <task_id>
+
+Then implement the task with these requirements:
+- Make the SMALLEST number of code changes possible
+- Follow ALL existing patterns in the codebase (you have access to analyze the code)
+- Do NOT over-engineer the solution
+- Use existing files/functions/patterns wherever possible
+- When complete, print: COMPLETED: <brief summary of changes>
+
+Begin by running tm show <task_id> to understand what needs to be implemented.
+```
+
+**Key References:**
+- [list.command.ts](mdc:apps/cli/src/commands/list.command.ts) - Command structure
+- [show.command.ts](mdc:apps/cli/src/commands/show.command.ts) - Task validation
+- Node.js `child_process.exec()` - For executing `claude "prompt"` command
+</PRD>

.taskmaster/state.json

@@ -1,6 +1,6 @@
 {
 	"currentTag": "master",
-	"lastSwitched": "2025-08-27T21:03:20.550Z",
+	"lastSwitched": "2025-09-12T22:25:27.535Z",
 	"branchTagMapping": {
 		"v017-adds": "v017-adds",
 		"next": "next"

.taskmaster/tasks/task_001_tm-start.txt

@@ -0,0 +1,34 @@
+# Task ID: 1
+# Title: Create start command class structure
+# Status: pending
+# Dependencies: None
+# Priority: high
+# Description: Create the basic structure for the start command following the Commander class pattern
+# Details:
+Create a new file `apps/cli/src/commands/start.command.ts` based on the existing list.command.ts pattern. Implement the command class with proper command registration, description, and argument handling for the task_id parameter. The class should extend the base Command class and implement the required methods.
+
+Example structure:
+```typescript
+import { Command } from 'commander';
+import { BaseCommand } from './base.command';
+
+export class StartCommand extends BaseCommand {
+  public register(program: Command): void {
+    program
+      .command('start')
+      .alias('tm start')
+      .description('Start implementing a task using claude-code')
+      .argument('<task_id>', 'ID of the task to start')
+      .action(async (taskId: string) => {
+        await this.execute(taskId);
+      });
+  }
+
+  public async execute(taskId: string): Promise<void> {
+    // Implementation will be added in subsequent tasks
+  }
+}
+```
+
+# Test Strategy:
+Verify the command registers correctly by running the CLI with --help and checking that the start command appears with proper description and arguments. Test the basic structure by ensuring the command can be invoked without errors.

.taskmaster/tasks/task_002_tm-start.txt

@@ -0,0 +1,26 @@
+# Task ID: 2
+# Title: Register start command in CLI
+# Status: pending
+# Dependencies: 7
+# Priority: high
+# Description: Register the start command in the CLI application
+# Details:
+Update the CLI application to register the new start command. This involves importing the StartCommand class and adding it to the commands array in the CLI initialization.
+
+In `apps/cli/src/index.ts` or the appropriate file where commands are registered:
+
+```typescript
+import { StartCommand } from './commands/start.command';
+
+// Add StartCommand to the commands array
+const commands = [
+  // ... existing commands
+  new StartCommand(),
+];
+
+// Register all commands
+commands.forEach(command => command.register(program));
+```
+
+# Test Strategy:
+Verify the command is correctly registered by running the CLI with --help and checking that the start command appears in the list of available commands.

.taskmaster/tasks/task_003_tm-start.txt

@@ -0,0 +1,32 @@
+# Task ID: 3
+# Title: Create standardized prompt builder
+# Status: pending
+# Dependencies: 1
+# Priority: medium
+# Description: Implement a function to build the standardized prompt for claude-code based on the task details
+# Details:
+Create a function in the StartCommand class that builds the standardized prompt according to the template provided in the PRD. The prompt should include instructions for Claude to first run `tm show <task_id>` to get task details, and then implement the required changes.
+
+```typescript
+private buildPrompt(taskId: string): string {
+  return `You are an AI coding assistant with access to this repository's codebase.
+
+First, run this command to get the task details:
+tm show ${taskId}
+
+Then implement the task with these requirements:
+- Make the SMALLEST number of code changes possible
+- Follow ALL existing patterns in the codebase (you have access to analyze the code)
+- Do NOT over-engineer the solution
+- Use existing files/functions/patterns wherever possible
+- When complete, print: COMPLETED: <brief summary of changes>
+
+Begin by running tm show ${taskId} to understand what needs to be implemented.`;
+}
+```
+<info added on 2025-09-12T02:40:01.812Z>
+The prompt builder function will handle task context retrieval by instructing Claude to use the task-master show command. This approach ensures Claude has access to all necessary task details before implementation begins. The command syntax "tm show ${taskId}" embedded in the prompt will direct Claude to first gather the complete task context, including description, requirements, and any existing implementation details, before proceeding with code changes.
+</info added on 2025-09-12T02:40:01.812Z>
+
+# Test Strategy:
+Verify the prompt is correctly formatted by calling the function with a sample task ID and checking that the output matches the expected template with the task ID properly inserted.

.taskmaster/tasks/task_004_tm-start.txt

@@ -0,0 +1,36 @@
+# Task ID: 4
+# Title: Implement claude-code executor
+# Status: pending
+# Dependencies: 3
+# Priority: high
+# Description: Add functionality to execute the claude-code command with the built prompt
+# Details:
+Implement the functionality to execute the claude command with the built prompt. This should use Node.js child_process.exec() to run the command directly in the terminal.
+
+```typescript
+import { exec } from 'child_process';
+
+// Inside execute method, after task validation
+private async executeClaude(prompt: string): Promise<void> {
+  console.log('Starting claude-code to implement the task...');
+  
+  try {
+    // Execute claude with the prompt
+    const claudeCommand = `claude "${prompt.replace(/"/g, '\\"')}"`;
+    
+    // Use execSync to wait for the command to complete
+    const { execSync } = require('child_process');
+    execSync(claudeCommand, { stdio: 'inherit' });
+    
+    console.log('Claude session completed.');
+  } catch (error) {
+    console.error('Error executing claude-code:', error.message);
+    process.exit(1);
+  }
+}
+```
+
+Then call this method from the execute method after building the prompt.
+
+# Test Strategy:
+Test by running the command with a valid task ID and verifying that the claude command is executed with the correct prompt. Check that the command handles errors appropriately if claude-code is not available.

.taskmaster/tasks/task_007_tm-start.txt

@@ -0,0 +1,49 @@
+# Task ID: 7
+# Title: Integrate execution flow in start command
+# Status: pending
+# Dependencies: 3, 4
+# Priority: high
+# Description: Connect all the components to implement the complete execution flow for the start command
+# Details:
+Update the execute method in the StartCommand class to integrate all the components and implement the complete execution flow as described in the PRD:
+1. Validate task exists
+2. Build standardized prompt
+3. Execute claude-code
+4. Check git status for changes
+5. Auto-mark task as done if changes detected
+
+```typescript
+public async execute(taskId: string): Promise<void> {
+  // Validate task exists
+  const core = await createTaskMasterCore();
+  const task = await core.tasks.getById(parseInt(taskId, 10));
+  
+  if (!task) {
+    console.error(`Task with ID ${taskId} not found`);
+    process.exit(1);
+  }
+  
+  // Build prompt
+  const prompt = this.buildPrompt(taskId);
+  
+  // Execute claude-code
+  await this.executeClaude(prompt);
+  
+  // Check git status
+  const changedFiles = await this.checkGitChanges();
+  
+  if (changedFiles.length > 0) {
+    console.log('\nChanges detected in the following files:');
+    changedFiles.forEach(file => console.log(`- ${file}`));
+    
+    // Auto-mark task as done
+    await this.markTaskAsDone(taskId);
+    console.log(`\nTask ${taskId} completed successfully and marked as done.`);
+  } else {
+    console.warn('\nNo changes detected after claude-code execution. Task not marked as done.');
+  }
+}
+```
+
+# Test Strategy:
+Test the complete execution flow by running the start command with a valid task ID and verifying that all steps are executed correctly. Test with both scenarios: when changes are detected and when no changes are detected.

CHANGELOG.md

@@ -1,5 +1,101 @@
 # task-master-ai
 
+## 0.27.2
+
+### Patch Changes
+
+- [#1248](https://github.com/eyaltoledano/claude-task-master/pull/1248) [`044a7bf`](https://github.com/eyaltoledano/claude-task-master/commit/044a7bfc98049298177bc655cf341d7a8b6a0011) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Fix set-status for subtasks:
+  - Parent tasks are now set as `done` when subtasks are all `done`
+  - Parent tasks are now set as `in-progress` when at least one subtask is `in-progress` or `done`
+
+## 0.27.1
+
+### Patch Changes
+
+- [#1232](https://github.com/eyaltoledano/claude-task-master/pull/1232) [`f487736`](https://github.com/eyaltoledano/claude-task-master/commit/f487736670ef8c484059f676293777eabb249c9e) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Fix module not found for new 0.27.0 release
+
+- [#1233](https://github.com/eyaltoledano/claude-task-master/pull/1233) [`c911608`](https://github.com/eyaltoledano/claude-task-master/commit/c911608f60454253f4e024b57ca84e5a5a53f65c) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Fix Zed MCP configuration by adding required "source" property
+  - Add "source": "custom" property to task-master-ai server in Zed settings.json
+
+## 0.27.1-rc.1
+
+### Patch Changes
+
+- [#1233](https://github.com/eyaltoledano/claude-task-master/pull/1233) [`1a18794`](https://github.com/eyaltoledano/claude-task-master/commit/1a1879483b86c118a4e46c02cbf4acebfcf6bcf9) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - One last testing final final
+
+## 0.27.1-rc.0
+
+### Patch Changes
+
+- [#1232](https://github.com/eyaltoledano/claude-task-master/pull/1232) [`f487736`](https://github.com/eyaltoledano/claude-task-master/commit/f487736670ef8c484059f676293777eabb249c9e) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Fix module not found for new 0.27.0 release
+
+## 0.27.0
+
+### Minor Changes
+
+- [#1220](https://github.com/eyaltoledano/claude-task-master/pull/1220) [`4e12643`](https://github.com/eyaltoledano/claude-task-master/commit/4e126430a092fb54afb035514fb3d46115714f97) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - No longer need --package=task-master-ai in mcp server
+  - A lot of users were having issues with Taskmaster and usually a simple fix was to remove --package from your mcp.json
+  - we now bundle our whole package, so we no longer need the --package
+
+- [#1200](https://github.com/eyaltoledano/claude-task-master/pull/1200) [`fce8414`](https://github.com/eyaltoledano/claude-task-master/commit/fce841490a9ebbf1801a42dd8a29397379cf1142) Thanks [@eyaltoledano](https://github.com/eyaltoledano)! - Add new `task-master start` command for automated task execution with Claude Code
+  - You can now start working on tasks directly by running `task-master start <task-id>` which will automatically launch Claude Code with a comprehensive prompt containing all task details, implementation guidelines, and context.
+  - `task-master start` will automatically detect next-task when no ID is provided.
+
+- [#1200](https://github.com/eyaltoledano/claude-task-master/pull/1200) [`fce8414`](https://github.com/eyaltoledano/claude-task-master/commit/fce841490a9ebbf1801a42dd8a29397379cf1142) Thanks [@eyaltoledano](https://github.com/eyaltoledano)! - Move from javascript to typescript, not a full refactor but we now have a typescript environment and are moving our javascript commands slowly into typescript
+
+- [#1200](https://github.com/eyaltoledano/claude-task-master/pull/1200) [`fce8414`](https://github.com/eyaltoledano/claude-task-master/commit/fce841490a9ebbf1801a42dd8a29397379cf1142) Thanks [@eyaltoledano](https://github.com/eyaltoledano)! - 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
+  1. **Get your Grok API key** from [console.x.ai](https://console.x.ai)
+  2. **Set the environment variable**:
+     ```bash
+     export GROK_CLI_API_KEY="your-api-key-here"
+     ```
+  3. **Configure Task Master to use Grok**:
+     ```bash
+     task-master models --set-main grok-beta
+     # or
+     task-master models --set-research grok-beta
+     # or
+     task-master models --set-fallback grok-beta
+     ```
+
+  ## 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
+
+  ## 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.
+
+- [#1225](https://github.com/eyaltoledano/claude-task-master/pull/1225) [`a621ff0`](https://github.com/eyaltoledano/claude-task-master/commit/a621ff05eafb51a147a9aabd7b37ddc0e45b0869) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - 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
+
+- [#1217](https://github.com/eyaltoledano/claude-task-master/pull/1217) [`e6de285`](https://github.com/eyaltoledano/claude-task-master/commit/e6de285ceacb0a397e952a63435cd32a9c731515) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - @tm/cli: add auto-update functionality to every command
+
+- [#1200](https://github.com/eyaltoledano/claude-task-master/pull/1200) [`fce8414`](https://github.com/eyaltoledano/claude-task-master/commit/fce841490a9ebbf1801a42dd8a29397379cf1142) Thanks [@eyaltoledano](https://github.com/eyaltoledano)! - Fix Grok model configuration validation and update deprecated Claude fallback model. Grok models now properly support their full 131K token capacity, and the fallback model has been upgraded to Claude Sonnet 4 for better performance and future compatibility.
+
+## 0.27.0-rc.2
+
+### Minor Changes
+
+- [#1217](https://github.com/eyaltoledano/claude-task-master/pull/1217) [`e6de285`](https://github.com/eyaltoledano/claude-task-master/commit/e6de285ceacb0a397e952a63435cd32a9c731515) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - @tm/cli: add auto-update functionality to every command
+
+## 0.27.0-rc.1
+
+### Minor Changes
+
+- [`255b9f0`](https://github.com/eyaltoledano/claude-task-master/commit/255b9f0334555b0063280abde701445cd62fa11b) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Testing one more pre-release iteration
+
 ## 0.27.0-rc.0
 
 ### Minor Changes

README.md

@@ -60,6 +60,19 @@ The following documentation is also available in the `docs` directory:
 
 > **Note:** After clicking the link, you'll still need to add your API keys to the configuration. The link installs the MCP server with placeholder keys that you'll need to replace with your actual API keys.
 
+#### Claude Code Quick Install
+
+For Claude Code users:
+
+```bash
+claude mcp add taskmaster-ai -- npx -y task-master-ai
+```
+
+Don't forget to add your API keys to the configuration:
+- in the root .env of your Project
+- in the "env" section of your mcp config for taskmaster-ai
+
+
 ## Requirements
 
 Taskmaster utilizes AI across several commands, and those require a separate API key. You can use a variety of models from different AI providers provided you add your API keys. For example, if you want to use Claude 3.7, you'll need an Anthropic API key.
@@ -92,17 +105,18 @@ MCP (Model Control Protocol) lets you run Task Master directly from your editor.
 |              | Project | `<project_folder>/.cursor/mcp.json`   | `<project_folder>\.cursor\mcp.json`               | `mcpServers` |
 | **Windsurf** | Global  | `~/.codeium/windsurf/mcp_config.json` | `%USERPROFILE%\.codeium\windsurf\mcp_config.json` | `mcpServers` |
 | **VS Code**  | Project | `<project_folder>/.vscode/mcp.json`   | `<project_folder>\.vscode\mcp.json`               | `servers`    |
+| **Q CLI**    | Global  | `~/.aws/amazonq/mcp.json`             |                                                   | `mcpServers` |
 
 ##### Manual Configuration
 
-###### Cursor & Windsurf (`mcpServers`)
+###### Cursor & Windsurf & Q Developer CLI (`mcpServers`)
 
 ```json
 {
   "mcpServers": {
     "task-master-ai": {
       "command": "npx",
-      "args": ["-y", "--package=task-master-ai", "task-master-ai"],
+      "args": ["-y", "task-master-ai"],
       "env": {
         "ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
         "PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY_HERE",
@@ -122,7 +136,7 @@ MCP (Model Control Protocol) lets you run Task Master directly from your editor.
 
 > 🔑 Replace `YOUR_…_KEY_HERE` with your real API keys. You can remove keys you don't use.
 
-> **Note**: If you see `0 tools enabled` in the MCP settings, try removing the `--package=task-master-ai` flag from `args`.
+> **Note**: If you see `0 tools enabled` in the MCP settings, restart your editor and check that your API keys are correctly configured.
 
 ###### VS Code (`servers` + `type`)
 
@@ -131,7 +145,7 @@ MCP (Model Control Protocol) lets you run Task Master directly from your editor.
   "servers": {
     "task-master-ai": {
       "command": "npx",
-      "args": ["-y", "--package=task-master-ai", "task-master-ai"],
+      "args": ["-y", "task-master-ai"],
       "env": {
         "ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
         "PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY_HERE",

apps/cli/CHANGELOG.md

@@ -1,5 +1,19 @@
 # @tm/cli
 
+## null
+
+### Patch Changes
+
+- Updated dependencies []:
+  - @tm/core@null
+
+## 0.27.0
+
+### Patch Changes
+
+- Updated dependencies []:
+  - @tm/core@0.26.1
+
 ## 0.27.0-rc.0
 
 ### Minor Changes

apps/cli/package.json

@@ -1,6 +1,5 @@
 {
 	"name": "@tm/cli",
-	"version": "0.27.0-rc.0",
 	"description": "Task Master CLI - Command line interface for task management",
 	"type": "module",
 	"private": true,
@@ -24,12 +23,12 @@
 	},
 	"dependencies": {
 		"@tm/core": "*",
-		"boxen": "^7.1.1",
+		"boxen": "^8.0.1",
 		"chalk": "5.6.2",
 		"cli-table3": "^0.6.5",
 		"commander": "^12.1.0",
-		"inquirer": "^9.2.10",
-		"ora": "^8.1.0"
+		"inquirer": "^12.5.0",
+		"ora": "^8.2.0"
 	},
 	"devDependencies": {
 		"@biomejs/biome": "^1.9.4",

apps/cli/src/commands/context.command.ts

@@ -6,7 +6,7 @@
 import { Command } from 'commander';
 import chalk from 'chalk';
 import inquirer from 'inquirer';
-import ora from 'ora';
+import ora, { Ora } from 'ora';
 import {
 	AuthManager,
 	AuthenticationError,
@@ -49,8 +49,15 @@ export class ContextCommand extends Command {
 		this.addClearCommand();
 		this.addSetCommand();
 
-		// Default action shows current context
-		this.action(async () => {
+		// Accept optional positional argument for brief ID or Hamster URL
+		this.argument('[briefOrUrl]', 'Brief ID or Hamster brief URL');
+
+		// Default action: if an argument is provided, resolve and set context; else show
+		this.action(async (briefOrUrl?: string) => {
+			if (briefOrUrl && briefOrUrl.trim().length > 0) {
+				await this.executeSetFromBriefInput(briefOrUrl.trim());
+				return;
+			}
 			await this.executeShow();
 		});
 	}
@@ -441,6 +448,142 @@ export class ContextCommand extends Command {
 		}
 	}
 
+	/**
+	 * Execute setting context from a brief ID or Hamster URL
+	 */
+	private async executeSetFromBriefInput(briefOrUrl: string): Promise<void> {
+		let spinner: Ora | undefined;
+		try {
+			// Check authentication
+			if (!this.authManager.isAuthenticated()) {
+				ui.displayError('Not authenticated. Run "tm auth login" first.');
+				process.exit(1);
+			}
+
+			spinner = ora('Resolving brief...');
+			spinner.start();
+
+			// Extract brief ID
+			const briefId = this.extractBriefId(briefOrUrl);
+			if (!briefId) {
+				spinner.fail('Could not extract a brief ID from the provided input');
+				ui.displayError(
+					`Provide a valid brief ID or a Hamster brief URL, e.g. https://${process.env.TM_PUBLIC_BASE_DOMAIN}/home/hamster/briefs/<id>`
+				);
+				process.exit(1);
+			}
+
+			// Fetch brief and resolve its organization
+			const brief = await this.authManager.getBrief(briefId);
+			if (!brief) {
+				spinner.fail('Brief not found or you do not have access');
+				process.exit(1);
+			}
+
+			// Fetch org to get a friendly name (optional)
+			let orgName: string | undefined;
+			try {
+				const org = await this.authManager.getOrganization(brief.accountId);
+				orgName = org?.name;
+			} catch {
+				// Non-fatal if org lookup fails
+			}
+
+			// Update context: set org and brief
+			const briefName = `Brief ${brief.id.slice(0, 8)}`;
+			await this.authManager.updateContext({
+				orgId: brief.accountId,
+				orgName,
+				briefId: brief.id,
+				briefName
+			});
+
+			spinner.succeed('Context set from brief');
+			console.log(
+				chalk.gray(
+					`  Organization: ${orgName || brief.accountId}\n  Brief: ${briefName}`
+				)
+			);
+
+			this.setLastResult({
+				success: true,
+				action: 'set',
+				context: this.authManager.getContext() || undefined,
+				message: 'Context set from brief'
+			});
+		} catch (error: any) {
+			try {
+				if (spinner?.isSpinning) spinner.stop();
+			} catch {}
+			this.handleError(error);
+			process.exit(1);
+		}
+	}
+
+	/**
+	 * Extract a brief ID from raw input (ID or Hamster URL)
+	 */
+	private extractBriefId(input: string): string | null {
+		const raw = input?.trim() ?? '';
+		if (!raw) return null;
+
+		const parseUrl = (s: string): URL | null => {
+			try {
+				return new URL(s);
+			} catch {}
+			try {
+				return new URL(`https://${s}`);
+			} catch {}
+			return null;
+		};
+
+		const fromParts = (path: string): string | null => {
+			const parts = path.split('/').filter(Boolean);
+			const briefsIdx = parts.lastIndexOf('briefs');
+			const candidate =
+				briefsIdx >= 0 && parts.length > briefsIdx + 1
+					? parts[briefsIdx + 1]
+					: parts[parts.length - 1];
+			return candidate?.trim() || null;
+		};
+
+		// 1) URL (absolute or scheme‑less)
+		const url = parseUrl(raw);
+		if (url) {
+			const qId = url.searchParams.get('id') || url.searchParams.get('briefId');
+			const candidate = (qId || fromParts(url.pathname)) ?? null;
+			if (candidate) {
+				// Light sanity check; let API be the final validator
+				if (this.isLikelyId(candidate) || candidate.length >= 8)
+					return candidate;
+			}
+		}
+
+		// 2) Looks like a path without scheme
+		if (raw.includes('/')) {
+			const candidate = fromParts(raw);
+			if (candidate && (this.isLikelyId(candidate) || candidate.length >= 8)) {
+				return candidate;
+			}
+		}
+
+		// 3) Fallback: raw token
+		return raw;
+	}
+
+	/**
+	 * Heuristic to check if a string looks like a brief ID (UUID-like)
+	 */
+	private isLikelyId(value: string): boolean {
+		const uuidRegex =
+			/^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{12}$/;
+		const ulidRegex = /^[0-9A-HJKMNP-TV-Z]{26}$/i; // ULID
+		const slugRegex = /^[A-Za-z0-9_-]{16,}$/; // general token
+		return (
+			uuidRegex.test(value) || ulidRegex.test(value) || slugRegex.test(value)
+		);
+	}
+
 	/**
 	 * Set context directly from options
 	 */

apps/cli/src/commands/show.command.ts

@@ -9,14 +9,7 @@ import boxen from 'boxen';
 import { createTaskMasterCore, type Task, type TaskMasterCore } from '@tm/core';
 import type { StorageType } from '@tm/core/types';
 import * as ui from '../utils/ui.js';
-import {
-	displayTaskHeader,
-	displayTaskProperties,
-	displayImplementationDetails,
-	displayTestStrategy,
-	displaySubtasks,
-	displaySuggestedActions
-} from '../ui/components/task-detail.component.js';
+import { displayTaskDetails } from '../ui/components/task-detail.component.js';
 
 /**
  * Options interface for the show command
@@ -264,44 +257,11 @@ export class ShowCommand extends Command {
 			return;
 		}
 
-		const task = result.task;
-
-		// Display header with tag
-		displayTaskHeader(task.id, task.title);
-
-		// Display task properties in table format
-		displayTaskProperties(task);
-
-		// Display implementation details if available
-		if (task.details) {
-			console.log(); // Empty line for spacing
-			displayImplementationDetails(task.details);
-		}
-
-		// Display test strategy if available
-		if ('testStrategy' in task && task.testStrategy) {
-			console.log(); // Empty line for spacing
-			displayTestStrategy(task.testStrategy as string);
-		}
-
-		// Display subtasks if available
-		if (task.subtasks && task.subtasks.length > 0) {
-			// Filter subtasks by status if provided
-			const filteredSubtasks = options.status
-				? task.subtasks.filter((sub) => sub.status === options.status)
-				: task.subtasks;
-
-			if (filteredSubtasks.length === 0 && options.status) {
-				console.log(
-					chalk.gray(`  No subtasks with status '${options.status}'`)
-				);
-			} else {
-				displaySubtasks(filteredSubtasks, task.id);
-			}
-		}
-
-		// Display suggested actions
-		displaySuggestedActions(task.id);
+		// Use the global task details display function
+		displayTaskDetails(result.task, {
+			statusFilter: options.status,
+			showSuggestedActions: true
+		});
 	}
 
 	/**

apps/cli/src/commands/start.command.ts

@@ -0,0 +1,512 @@
+/**
+ * @fileoverview StartCommand using Commander's native class pattern
+ * Extends Commander.Command for better integration with the framework
+ * This is a thin presentation layer over @tm/core's TaskExecutionService
+ */
+
+import { Command } from 'commander';
+import chalk from 'chalk';
+import boxen from 'boxen';
+import ora, { type Ora } from 'ora';
+import { spawn } from 'child_process';
+import {
+	createTaskMasterCore,
+	type TaskMasterCore,
+	type StartTaskResult as CoreStartTaskResult
+} from '@tm/core';
+import { displayTaskDetails } from '../ui/components/task-detail.component.js';
+import * as ui from '../utils/ui.js';
+
+/**
+ * CLI-specific options interface for the start command
+ */
+export interface StartCommandOptions {
+	id?: string;
+	format?: 'text' | 'json';
+	project?: string;
+	dryRun?: boolean;
+	force?: boolean;
+	noStatusUpdate?: boolean;
+}
+
+/**
+ * CLI-specific result type from start command
+ * Extends the core result with CLI-specific display information
+ */
+export interface StartCommandResult extends CoreStartTaskResult {
+	storageType?: string;
+}
+
+/**
+ * StartCommand extending Commander's Command class
+ * This is a thin presentation layer over @tm/core's TaskExecutionService
+ */
+export class StartCommand extends Command {
+	private tmCore?: TaskMasterCore;
+	private lastResult?: StartCommandResult;
+
+	constructor(name?: string) {
+		super(name || 'start');
+
+		// Configure the command
+		this.description(
+			'Start working on a task by launching claude-code with context'
+		)
+			.argument('[id]', 'Task ID to start working on')
+			.option('-i, --id <id>', 'Task ID to start working on')
+			.option('-f, --format <format>', 'Output format (text, json)', 'text')
+			.option('-p, --project <path>', 'Project root directory', process.cwd())
+			.option(
+				'--dry-run',
+				'Show what would be executed without launching claude-code'
+			)
+			.option(
+				'--force',
+				'Force start even if another task is already in-progress'
+			)
+			.option(
+				'--no-status-update',
+				'Do not automatically update task status to in-progress'
+			)
+			.action(
+				async (taskId: string | undefined, options: StartCommandOptions) => {
+					await this.executeCommand(taskId, options);
+				}
+			);
+	}
+
+	/**
+	 * Execute the start command
+	 */
+	private async executeCommand(
+		taskId: string | undefined,
+		options: StartCommandOptions
+	): Promise<void> {
+		let spinner: Ora | null = null;
+
+		try {
+			// Validate options
+			if (!this.validateOptions(options)) {
+				process.exit(1);
+			}
+
+			// Initialize tm-core with spinner
+			spinner = ora('Initializing Task Master...').start();
+			await this.initializeCore(options.project || process.cwd());
+			spinner.succeed('Task Master initialized');
+
+			// Get the task ID from argument or option, or find next available task
+			const idArg = taskId || options.id || null;
+			let targetTaskId = idArg;
+
+			if (!targetTaskId) {
+				spinner = ora('Finding next available task...').start();
+				targetTaskId = await this.performGetNextTask();
+				if (targetTaskId) {
+					spinner.succeed(`Found next task: #${targetTaskId}`);
+				} else {
+					spinner.fail('No available tasks found');
+				}
+			}
+
+			if (!targetTaskId) {
+				ui.displayError('No task ID provided and no available tasks found');
+				process.exit(1);
+			}
+
+			// Show pre-launch message (no spinner needed, it's just display)
+			if (!options.dryRun) {
+				await this.showPreLaunchMessage(targetTaskId);
+			}
+
+			// Use tm-core's startTask method with spinner
+			spinner = ora('Preparing task execution...').start();
+			const coreResult = await this.performStartTask(targetTaskId, options);
+
+			if (coreResult.started) {
+				spinner.succeed(
+					options.dryRun
+						? 'Dry run completed'
+						: 'Task prepared - launching Claude...'
+				);
+			} else {
+				spinner.fail('Task execution failed');
+			}
+
+			// Execute command if we have one and it's not a dry run
+			if (!options.dryRun && coreResult.command) {
+				// Stop any remaining spinners before launching Claude
+				if (spinner && !spinner.isSpinning) {
+					// Clear the line to make room for Claude
+					console.log();
+				}
+				await this.executeChildProcess(coreResult.command);
+			}
+
+			// Convert core result to CLI result with storage type
+			const result: StartCommandResult = {
+				...coreResult,
+				storageType: this.tmCore?.getStorageType()
+			};
+
+			// Store result for programmatic access
+			this.setLastResult(result);
+
+			// Display results (only for dry run or if execution failed)
+			if (options.dryRun || !coreResult.started) {
+				this.displayResults(result, options);
+			}
+		} catch (error: any) {
+			if (spinner) {
+				spinner.fail('Operation failed');
+			}
+			this.handleError(error);
+			process.exit(1);
+		}
+	}
+
+	/**
+	 * Validate command options
+	 */
+	private validateOptions(options: StartCommandOptions): boolean {
+		// Validate format
+		if (options.format && !['text', 'json'].includes(options.format)) {
+			console.error(chalk.red(`Invalid format: ${options.format}`));
+			console.error(chalk.gray(`Valid formats: text, json`));
+			return false;
+		}
+
+		return true;
+	}
+
+	/**
+	 * Initialize TaskMasterCore
+	 */
+	private async initializeCore(projectRoot: string): Promise<void> {
+		if (!this.tmCore) {
+			this.tmCore = await createTaskMasterCore({ projectPath: projectRoot });
+		}
+	}
+
+	/**
+	 * Get the next available task using tm-core
+	 */
+	private async performGetNextTask(): Promise<string | null> {
+		if (!this.tmCore) {
+			throw new Error('TaskMasterCore not initialized');
+		}
+		return this.tmCore.getNextAvailableTask();
+	}
+
+	/**
+	 * Show pre-launch message using tm-core data
+	 */
+	private async showPreLaunchMessage(targetTaskId: string): Promise<void> {
+		if (!this.tmCore) return;
+
+		const { task, subtask, subtaskId } =
+			await this.tmCore.getTaskWithSubtask(targetTaskId);
+		if (task) {
+			const workItemText = subtask
+				? `Subtask #${task.id}.${subtaskId} - ${subtask.title}`
+				: `Task #${task.id} - ${task.title}`;
+
+			console.log(
+				chalk.green('🚀 Starting: ') + chalk.white.bold(workItemText)
+			);
+			console.log(chalk.gray('Launching Claude Code...'));
+			console.log(); // Empty line
+		}
+	}
+
+	/**
+	 * Perform start task using tm-core business logic
+	 */
+	private async performStartTask(
+		targetTaskId: string,
+		options: StartCommandOptions
+	): Promise<CoreStartTaskResult> {
+		if (!this.tmCore) {
+			throw new Error('TaskMasterCore not initialized');
+		}
+
+		// Show spinner for status update if enabled
+		let statusSpinner: Ora | null = null;
+		if (!options.noStatusUpdate && !options.dryRun) {
+			statusSpinner = ora('Updating task status to in-progress...').start();
+		}
+
+		// Get execution command from tm-core (instead of executing directly)
+		const result = await this.tmCore.startTask(targetTaskId, {
+			dryRun: options.dryRun,
+			force: options.force,
+			updateStatus: !options.noStatusUpdate
+		});
+
+		if (statusSpinner) {
+			if (result.started) {
+				statusSpinner.succeed('Task status updated');
+			} else {
+				statusSpinner.warn('Task status update skipped');
+			}
+		}
+
+		if (!result) {
+			throw new Error('Failed to start task - core result is undefined');
+		}
+
+		// Don't execute here - let the main executeCommand method handle it
+		return result;
+	}
+
+	/**
+	 * Execute the child process directly in the main thread for better process control
+	 */
+	private async executeChildProcess(command: {
+		executable: string;
+		args: string[];
+		cwd: string;
+	}): Promise<void> {
+		return new Promise((resolve, reject) => {
+			// Don't show the full command with args as it can be very long
+			console.log(chalk.green('🚀 Launching Claude Code...'));
+			console.log(); // Add space before Claude takes over
+
+			const childProcess = spawn(command.executable, command.args, {
+				cwd: command.cwd,
+				stdio: 'inherit', // Inherit stdio from parent process
+				shell: false
+			});
+
+			childProcess.on('close', (code) => {
+				if (code === 0) {
+					resolve();
+				} else {
+					reject(new Error(`Process exited with code ${code}`));
+				}
+			});
+
+			childProcess.on('error', (error) => {
+				reject(new Error(`Failed to spawn process: ${error.message}`));
+			});
+
+			// Handle process termination signals gracefully
+			const cleanup = () => {
+				if (childProcess && !childProcess.killed) {
+					childProcess.kill('SIGTERM');
+				}
+			};
+
+			process.on('SIGINT', cleanup);
+			process.on('SIGTERM', cleanup);
+			process.on('exit', cleanup);
+		});
+	}
+
+	/**
+	 * Display results based on format
+	 */
+	private displayResults(
+		result: StartCommandResult,
+		options: StartCommandOptions
+	): void {
+		const format = options.format || 'text';
+
+		switch (format) {
+			case 'json':
+				this.displayJson(result);
+				break;
+
+			case 'text':
+			default:
+				this.displayTextResult(result, options);
+				break;
+		}
+	}
+
+	/**
+	 * Display in JSON format
+	 */
+	private displayJson(result: StartCommandResult): void {
+		console.log(JSON.stringify(result, null, 2));
+	}
+
+	/**
+	 * Display result in text format
+	 */
+	private displayTextResult(
+		result: StartCommandResult,
+		options: StartCommandOptions
+	): void {
+		if (!result.found || !result.task) {
+			console.log(
+				boxen(chalk.yellow(`Task not found!`), {
+					padding: { top: 0, bottom: 0, left: 1, right: 1 },
+					borderColor: 'yellow',
+					borderStyle: 'round',
+					margin: { top: 1 }
+				})
+			);
+			return;
+		}
+
+		const task = result.task;
+
+		if (options.dryRun) {
+			// For dry run, show full details since Claude Code won't be launched
+			let headerText = `Dry Run: Starting Task #${task.id} - ${task.title}`;
+
+			// If working on a specific subtask, highlight it in the header
+			if (result.subtask && result.subtaskId) {
+				headerText = `Dry Run: Starting Subtask #${task.id}.${result.subtaskId} - ${result.subtask.title}`;
+			}
+
+			displayTaskDetails(task, {
+				customHeader: headerText,
+				headerColor: 'yellow'
+			});
+
+			// Show claude-code prompt
+			if (result.executionOutput) {
+				console.log(); // Empty line for spacing
+				console.log(
+					boxen(
+						chalk.white.bold('Claude-Code Prompt:') +
+							'\n\n' +
+							result.executionOutput,
+						{
+							padding: 1,
+							borderStyle: 'round',
+							borderColor: 'cyan',
+							width: process.stdout.columns * 0.95 || 100
+						}
+					)
+				);
+			}
+
+			console.log(); // Empty line for spacing
+			console.log(
+				boxen(
+					chalk.yellow(
+						'🔍 Dry run - claude-code would be launched with the above prompt'
+					),
+					{
+						padding: { top: 0, bottom: 0, left: 1, right: 1 },
+						borderColor: 'yellow',
+						borderStyle: 'round'
+					}
+				)
+			);
+		} else {
+			// For actual execution, show minimal info since Claude Code will clear the terminal
+			if (result.started) {
+				// Determine what was worked on - task or subtask
+				let workItemText = `Task: #${task.id} - ${task.title}`;
+				let statusTarget = task.id;
+
+				if (result.subtask && result.subtaskId) {
+					workItemText = `Subtask: #${task.id}.${result.subtaskId} - ${result.subtask.title}`;
+					statusTarget = `${task.id}.${result.subtaskId}`;
+				}
+
+				// Post-execution message (shown after Claude Code exits)
+				console.log(
+					boxen(
+						chalk.green.bold('🎉 Task Session Complete!') +
+							'\n\n' +
+							chalk.white(workItemText) +
+							'\n\n' +
+							chalk.cyan('Next steps:') +
+							'\n' +
+							`• Run ${chalk.yellow('tm show ' + task.id)} to review task details\n` +
+							`• Run ${chalk.yellow('tm set-status --id=' + statusTarget + ' --status=done')} when complete\n` +
+							`• Run ${chalk.yellow('tm next')} to find the next available task\n` +
+							`• Run ${chalk.yellow('tm start')} to begin the next task`,
+						{
+							padding: 1,
+							borderStyle: 'round',
+							borderColor: 'green',
+							width: process.stdout.columns * 0.95 || 100,
+							margin: { top: 1 }
+						}
+					)
+				);
+			} else {
+				// Error case
+				console.log(
+					boxen(
+						chalk.red(
+							'❌ Failed to launch claude-code' +
+								(result.error ? `\nError: ${result.error}` : '')
+						),
+						{
+							padding: { top: 0, bottom: 0, left: 1, right: 1 },
+							borderColor: 'red',
+							borderStyle: 'round'
+						}
+					)
+				);
+			}
+		}
+
+		console.log(`\n${chalk.gray('Storage: ' + result.storageType)}`);
+	}
+
+	/**
+	 * Handle general errors
+	 */
+	private handleError(error: any): void {
+		const msg = error?.getSanitizedDetails?.() ?? {
+			message: error?.message ?? String(error)
+		};
+		console.error(chalk.red(`Error: ${msg.message || 'Unexpected error'}`));
+
+		// Show stack trace in development mode or when DEBUG is set
+		const isDevelopment = process.env.NODE_ENV !== 'production';
+		if ((isDevelopment || process.env.DEBUG) && error.stack) {
+			console.error(chalk.gray(error.stack));
+		}
+	}
+
+	/**
+	 * Set the last result for programmatic access
+	 */
+	private setLastResult(result: StartCommandResult): void {
+		this.lastResult = result;
+	}
+
+	/**
+	 * Get the last result (for programmatic usage)
+	 */
+	getLastResult(): StartCommandResult | undefined {
+		return this.lastResult;
+	}
+
+	/**
+	 * Clean up resources
+	 */
+	async cleanup(): Promise<void> {
+		if (this.tmCore) {
+			await this.tmCore.close();
+			this.tmCore = undefined;
+		}
+	}
+
+	/**
+	 * Static method to register this command on an existing program
+	 */
+	static registerOn(program: Command): Command {
+		const startCommand = new StartCommand();
+		program.addCommand(startCommand);
+		return startCommand;
+	}
+
+	/**
+	 * Alternative registration that returns the command for chaining
+	 */
+	static register(program: Command, name?: string): StartCommand {
+		const startCommand = new StartCommand(name);
+		program.addCommand(startCommand);
+		return startCommand;
+	}
+}

apps/cli/src/index.ts

@@ -8,11 +8,20 @@ export { ListTasksCommand } from './commands/list.command.js';
 export { ShowCommand } from './commands/show.command.js';
 export { AuthCommand } from './commands/auth.command.js';
 export { ContextCommand } from './commands/context.command.js';
+export { StartCommand } from './commands/start.command.js';
 export { SetStatusCommand } from './commands/set-status.command.js';
 
 // UI utilities (for other commands to use)
 export * as ui from './utils/ui.js';
 
+// Auto-update utilities
+export {
+	checkForUpdate,
+	performAutoUpdate,
+	displayUpgradeNotification,
+	compareVersions
+} from './utils/auto-update.js';
+
 // Re-export commonly used types from tm-core
 export type {
 	Task,

apps/cli/src/ui/components/task-detail.component.ts

@@ -262,3 +262,74 @@ export function displaySuggestedActions(taskId: string | number): void {
 		)
 	);
 }
+
+/**
+ * Display complete task details - used by both show and start commands
+ */
+export function displayTaskDetails(
+	task: Task,
+	options?: {
+		statusFilter?: string;
+		showSuggestedActions?: boolean;
+		customHeader?: string;
+		headerColor?: string;
+	}
+): void {
+	const {
+		statusFilter,
+		showSuggestedActions = false,
+		customHeader,
+		headerColor = 'blue'
+	} = options || {};
+
+	// Display header - either custom or default
+	if (customHeader) {
+		console.log(
+			boxen(chalk.white.bold(customHeader), {
+				padding: { top: 0, bottom: 0, left: 1, right: 1 },
+				borderColor: headerColor,
+				borderStyle: 'round',
+				margin: { top: 1 }
+			})
+		);
+	} else {
+		displayTaskHeader(task.id, task.title);
+	}
+
+	// Display task properties in table format
+	displayTaskProperties(task);
+
+	// Display implementation details if available
+	if (task.details) {
+		console.log(); // Empty line for spacing
+		displayImplementationDetails(task.details);
+	}
+
+	// Display test strategy if available
+	if ('testStrategy' in task && task.testStrategy) {
+		console.log(); // Empty line for spacing
+		displayTestStrategy(task.testStrategy as string);
+	}
+
+	// Display subtasks if available
+	if (task.subtasks && task.subtasks.length > 0) {
+		// Filter subtasks by status if provided
+		const filteredSubtasks = statusFilter
+			? task.subtasks.filter((sub) => sub.status === statusFilter)
+			: task.subtasks;
+
+		if (filteredSubtasks.length === 0 && statusFilter) {
+			console.log(); // Empty line for spacing
+			console.log(chalk.gray(`  No subtasks with status '${statusFilter}'`));
+		} else if (filteredSubtasks.length > 0) {
+			console.log(); // Empty line for spacing
+			displaySubtasks(filteredSubtasks, task.id);
+		}
+	}
+
+	// Display suggested actions if requested
+	if (showSuggestedActions) {
+		console.log(); // Empty line for spacing
+		displaySuggestedActions(task.id);
+	}
+}

apps/cli/src/utils/auto-update.ts

@@ -0,0 +1,240 @@
+/**
+ * @fileoverview Auto-update utilities for task-master-ai CLI
+ */
+
+import { spawn } from 'child_process';
+import https from 'https';
+import chalk from 'chalk';
+import ora from 'ora';
+import boxen from 'boxen';
+
+export interface UpdateInfo {
+	currentVersion: string;
+	latestVersion: string;
+	needsUpdate: boolean;
+}
+
+/**
+ * Get current version from build-time injected environment variable
+ */
+function getCurrentVersion(): string {
+	// Version is injected at build time via TM_PUBLIC_VERSION
+	const version = process.env.TM_PUBLIC_VERSION;
+	if (version && version !== 'unknown') {
+		return version;
+	}
+
+	// Fallback for development or if injection failed
+	console.warn('Could not read version from TM_PUBLIC_VERSION, using fallback');
+	return '0.0.0';
+}
+
+/**
+ * Compare semantic versions with proper pre-release handling
+ * @param v1 - First version
+ * @param v2 - Second version
+ * @returns -1 if v1 < v2, 0 if v1 = v2, 1 if v1 > v2
+ */
+export function compareVersions(v1: string, v2: string): number {
+	const toParts = (v: string) => {
+		const [core, pre = ''] = v.split('-', 2);
+		const nums = core.split('.').map((n) => Number.parseInt(n, 10) || 0);
+		return { nums, pre };
+	};
+
+	const a = toParts(v1);
+	const b = toParts(v2);
+	const len = Math.max(a.nums.length, b.nums.length);
+
+	// Compare numeric parts
+	for (let i = 0; i < len; i++) {
+		const d = (a.nums[i] || 0) - (b.nums[i] || 0);
+		if (d !== 0) return d < 0 ? -1 : 1;
+	}
+
+	// Handle pre-release comparison
+	if (a.pre && !b.pre) return -1; // prerelease < release
+	if (!a.pre && b.pre) return 1; // release > prerelease
+	if (a.pre === b.pre) return 0; // same or both empty
+	return a.pre < b.pre ? -1 : 1; // basic prerelease tie-break
+}
+
+/**
+ * Check for newer version of task-master-ai
+ */
+export async function checkForUpdate(
+	currentVersionOverride?: string
+): Promise<UpdateInfo> {
+	const currentVersion = currentVersionOverride || getCurrentVersion();
+
+	return new Promise((resolve) => {
+		const options = {
+			hostname: 'registry.npmjs.org',
+			path: '/task-master-ai',
+			method: 'GET',
+			headers: {
+				Accept: 'application/vnd.npm.install-v1+json',
+				'User-Agent': `task-master-ai/${currentVersion}`
+			}
+		};
+
+		const req = https.request(options, (res) => {
+			let data = '';
+
+			res.on('data', (chunk) => {
+				data += chunk;
+			});
+
+			res.on('end', () => {
+				try {
+					if (res.statusCode !== 200)
+						throw new Error(`npm registry status ${res.statusCode}`);
+					const npmData = JSON.parse(data);
+					const latestVersion = npmData['dist-tags']?.latest || currentVersion;
+
+					const needsUpdate =
+						compareVersions(currentVersion, latestVersion) < 0;
+
+					resolve({
+						currentVersion,
+						latestVersion,
+						needsUpdate
+					});
+				} catch (error) {
+					resolve({
+						currentVersion,
+						latestVersion: currentVersion,
+						needsUpdate: false
+					});
+				}
+			});
+		});
+
+		req.on('error', () => {
+			resolve({
+				currentVersion,
+				latestVersion: currentVersion,
+				needsUpdate: false
+			});
+		});
+
+		req.setTimeout(3000, () => {
+			req.destroy();
+			resolve({
+				currentVersion,
+				latestVersion: currentVersion,
+				needsUpdate: false
+			});
+		});
+
+		req.end();
+	});
+}
+
+/**
+ * Display upgrade notification message
+ */
+export function displayUpgradeNotification(
+	currentVersion: string,
+	latestVersion: string
+) {
+	const message = boxen(
+		`${chalk.blue.bold('Update Available!')} ${chalk.dim(currentVersion)} → ${chalk.green(latestVersion)}\n\n` +
+			`Auto-updating to the latest version with new features and bug fixes...`,
+		{
+			padding: 1,
+			margin: { top: 1, bottom: 1 },
+			borderColor: 'yellow',
+			borderStyle: 'round'
+		}
+	);
+
+	console.log(message);
+}
+
+/**
+ * Automatically update task-master-ai to the latest version
+ */
+export async function performAutoUpdate(
+	latestVersion: string
+): Promise<boolean> {
+	if (process.env.TASKMASTER_SKIP_AUTO_UPDATE === '1' || process.env.CI) {
+		console.log(
+			chalk.dim('Skipping auto-update (TASKMASTER_SKIP_AUTO_UPDATE/CI).')
+		);
+		return false;
+	}
+	const spinner = ora({
+		text: chalk.blue(
+			`Updating task-master-ai to version ${chalk.green(latestVersion)}`
+		),
+		spinner: 'dots',
+		color: 'blue'
+	}).start();
+
+	return new Promise((resolve) => {
+		const updateProcess = spawn(
+			'npm',
+			[
+				'install',
+				'-g',
+				`task-master-ai@${latestVersion}`,
+				'--no-fund',
+				'--no-audit',
+				'--loglevel=warn'
+			],
+			{
+				stdio: ['ignore', 'pipe', 'pipe']
+			}
+		);
+
+		let errorOutput = '';
+
+		updateProcess.stdout.on('data', () => {
+			// Update spinner text with progress
+			spinner.text = chalk.blue(
+				`Installing task-master-ai@${latestVersion}...`
+			);
+		});
+
+		updateProcess.stderr.on('data', (data) => {
+			errorOutput += data.toString();
+		});
+
+		updateProcess.on('close', (code) => {
+			if (code === 0) {
+				spinner.succeed(
+					chalk.green(
+						`Successfully updated to version ${chalk.bold(latestVersion)}`
+					)
+				);
+				console.log(
+					chalk.dim('Please restart your command to use the new version.')
+				);
+				resolve(true);
+			} else {
+				spinner.fail(chalk.red('Auto-update failed'));
+				console.log(
+					chalk.cyan(
+						`Please run manually: npm install -g task-master-ai@${latestVersion}`
+					)
+				);
+				if (errorOutput) {
+					console.log(chalk.dim(`Error: ${errorOutput.trim()}`));
+				}
+				resolve(false);
+			}
+		});
+
+		updateProcess.on('error', (error) => {
+			spinner.fail(chalk.red('Auto-update failed'));
+			console.log(chalk.red('Error:'), error.message);
+			console.log(
+				chalk.cyan(
+					`Please run manually: npm install -g task-master-ai@${latestVersion}`
+				)
+			);
+			resolve(false);
+		});
+	});
+}

apps/docs/CHANGELOG.md

@@ -1,5 +1,9 @@
 # docs
 
+## 0.0.4
+
+## 0.0.3
+
 ## 0.0.2
 
 ## 0.0.1

apps/docs/README.md

@@ -1,22 +1,24 @@
 # Task Master Documentation
 
-Welcome to the Task Master documentation. Use the links below to navigate to the information you need:
+Welcome to the Task Master documentation. This documentation site provides comprehensive guides for getting started with Task Master.
 
 ## Getting Started
 
-- [Configuration Guide](archive/configuration.md) - Set up environment variables and customize Task Master
-- [Tutorial](archive/ctutorial.md) - Step-by-step guide to getting started with Task Master
+- [Quick Start Guide](/getting-started/quick-start) - Complete setup and first-time usage guide
+- [Requirements](/getting-started/quick-start/requirements) - What you need to get started
+- [Installation](/getting-started/quick-start/installation) - How to install Task Master
 
-## Reference
+## Core Capabilities
 
-- [Command Reference](archive/ccommand-reference.md) - Complete list of all available commands
-- [Task Structure](archive/ctask-structure.md) - Understanding the task format and features
+- [MCP Tools](/capabilities/mcp) - Model Control Protocol integration
+- [CLI Commands](/capabilities/cli-root-commands) - Command line interface reference
+- [Task Structure](/capabilities/task-structure) - Understanding tasks and subtasks
 
-## Examples & Licensing
+## Best Practices
 
-- [Example Interactions](archive/cexamples.md) - Common Cursor AI interaction examples
-- [Licensing Information](archive/clicensing.md) - Detailed information about the license
+- [Advanced Configuration](/best-practices/configuration-advanced) - Detailed configuration options
+- [Advanced Tasks](/best-practices/advanced-tasks) - Working with complex task structures
 
 ## Need More Help?
 
-If you can't find what you're looking for in these docs, please check the [main README](../README.md) or visit our [GitHub repository](https://github.com/eyaltoledano/claude-task-master).
+If you can't find what you're looking for in these docs, please check the root README.md or visit our [GitHub repository](https://github.com/eyaltoledano/claude-task-master).

apps/docs/archive/Installation.mdx

@@ -19,7 +19,7 @@ description: "This guide walks you through setting up Task Master in your develo
         	"mcpServers": {
         		"taskmaster-ai": {
         			"command": "npx",
-        			"args": ["-y", "--package", "task-master-ai", "task-master-mcp"],
+        			"args": ["-y", "task-master-ai"],
         			"env": {
         				"ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
         				"PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY_HERE",

apps/docs/archive/cursor-setup.mdx

@@ -48,7 +48,7 @@ description: "Learn how to set up and use Task Master with Cursor AI"
       <Step title="Configure with the following details:">
         - Name: "Task Master"
         - Type: "Command"
-        - Command: "npx -y --package task-master-ai task-master-mcp"
+        - Command: "npx -y task-master-ai"
       </Step>
       <Step title="Save Settings">
         

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

@@ -83,6 +83,8 @@ Taskmaster uses two primary methods for configuration:
   - `VERTEX_PROJECT_ID`: Your Google Cloud project ID for Vertex AI. Required when using the 'vertex' provider.
   - `VERTEX_LOCATION`: Google Cloud region for Vertex AI (e.g., 'us-central1'). Default is 'us-central1'.
   - `GOOGLE_APPLICATION_CREDENTIALS`: Path to service account credentials JSON file for Google Cloud auth (alternative to API key for Vertex AI).
+- **Optional Auto-Update Control:**
+  - `TASKMASTER_SKIP_AUTO_UPDATE`: Set to '1' to disable automatic updates. Also automatically disabled in CI environments (when `CI` environment variable is set).
 
 **Important:** Settings like model ID selections (`main`, `research`, `fallback`), `maxTokens`, `temperature`, `logLevel`, `defaultSubtasks`, `defaultPriority`, and `projectName` are **managed in `.taskmaster/config.json`** (or `.taskmasterconfig` for unmigrated projects), not environment variables.
 

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

@@ -156,7 +156,7 @@ sidebarTitle: "CLI Commands"
     # Use an alternative tasks file
     task-master analyze-complexity --file=custom-tasks.json
 
-    # Use Perplexity AI for research-backed complexity analysis
+    # Use your configured research model for research-backed complexity analysis
     task-master analyze-complexity --research
     ```
   </Accordion>

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

@@ -18,8 +18,8 @@ For MCP/Cursor usage: Configure keys in the env section of your .cursor/mcp.json
 {
 	"mcpServers": {
 		"task-master-ai": {
-			"command": "node",
-			"args": ["./mcp-server/server.js"],
+			"command": "npx",
+			"args": ["-y", "task-master-ai"],
 			"env": {
 				"ANTHROPIC_API_KEY": "ANTHROPIC_API_KEY_HERE",
 				"PERPLEXITY_API_KEY": "PERPLEXITY_API_KEY_HERE",
@@ -108,5 +108,5 @@ You don’t need to configure everything up front. Most settings can be left as
 </Accordion>
 
 <Note>
-For advanced configuration options and detailed customization, see our [Advanced Configuration Guide](/docs/best-practices/configuration-advanced) page.
+For advanced configuration options and detailed customization, see our [Advanced Configuration Guide](/best-practices/configuration-advanced) page.
 </Note>

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

@@ -56,4 +56,4 @@ If you ran into problems and had to debug errors you can create new rules as you
 
 By now you have all you need to get started executing code faster and smarter with Task Master.
 
-If you have any questions please check out [Frequently Asked Questions](/docs/getting-started/faq)
+If you have any questions please check out [Frequently Asked Questions](/getting-started/faq)

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

@@ -30,6 +30,19 @@ cursor://anysphere.cursor-deeplink/mcp/install?name=taskmaster-ai&config=eyJjb21
 ```
 
 > **Note:** After clicking the link, you'll still need to add your API keys to the configuration. The link installs the MCP server with placeholder keys that you'll need to replace with your actual API keys.
+
+### Claude Code Quick Install
+
+For Claude Code users:
+
+```bash
+claude mcp add taskmaster-ai -- npx -y task-master-ai
+```
+
+Don't forget to add your API keys to the configuration:
+- in the root .env of your Project
+- in the "env" section of your mcp config for taskmaster-ai
+
 </Accordion>
 ## Installation Options
 
@@ -56,7 +69,7 @@ MCP (Model Control Protocol) lets you run Task Master directly from your editor.
   "mcpServers": {
     "taskmaster-ai": {
       "command": "npx",
-      "args": ["-y", "--package=task-master-ai", "task-master-ai"],
+      "args": ["-y", "task-master-ai"],
       "env": {
         "ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
         "PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY_HERE",
@@ -75,7 +88,7 @@ MCP (Model Control Protocol) lets you run Task Master directly from your editor.
 
 > 🔑 Replace `YOUR_…_KEY_HERE` with your real API keys. You can remove keys you don't use.
 
-> **Note**: If you see `0 tools enabled` in the MCP settings, try removing the `--package=task-master-ai` flag from `args`.
+> **Note**: If you see `0 tools enabled` in the MCP settings, restart your editor and check that your API keys are correctly configured.
 
 ### VS Code (`servers` + `type`)
 
@@ -84,7 +97,7 @@ MCP (Model Control Protocol) lets you run Task Master directly from your editor.
   "servers": {
     "taskmaster-ai": {
       "command": "npx",
-      "args": ["-y", "--package=task-master-ai", "task-master-ai"],
+      "args": ["-y", "task-master-ai"],
       "env": {
         "ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
         "PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY_HERE",

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

@@ -6,13 +6,13 @@ sidebarTitle: "Quick Start"
 This guide is for new users who want to start using Task Master with minimal setup time.
 
 It covers:
-- [Requirements](/docs/getting-started/quick-start/requirements): You will need Node.js and an AI model API Key.
-- [Installation](/docs/getting-started/quick-start/installation): How to Install Task Master.
-- [Configuration](/docs/getting-started/quick-start/configuration-quick): Setting up your API Key, MCP, and more.
-- [PRD](/docs/getting-started/quick-start/prd-quick): Writing and parsing your first PRD.
-- [Task Setup](/docs/getting-started/quick-start/tasks-quick): Preparing your tasks for execution.
-- [Executing Tasks](/docs/getting-started/quick-start/execute-quick): Using Task Master to execute tasks.
-- [Rules & Context](/docs/getting-started/quick-start/rules-quick): Learn how and why to build context in your project over time.
+- [Requirements](/getting-started/quick-start/requirements): You will need Node.js and an AI model API Key.
+- [Installation](/getting-started/quick-start/installation): How to Install Task Master.
+- [Configuration](/getting-started/quick-start/configuration-quick): Setting up your API Key, MCP, and more.
+- [PRD](/getting-started/quick-start/prd-quick): Writing and parsing your first PRD.
+- [Task Setup](/getting-started/quick-start/tasks-quick): Preparing your tasks for execution.
+- [Executing Tasks](/getting-started/quick-start/execute-quick): Using Task Master to execute tasks.
+- [Rules & Context](/getting-started/quick-start/rules-quick): Learn how and why to build context in your project over time.
 
 <Tip>
 By the end of this guide, you'll have everything you need to begin working productively with Task Master.

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

@@ -61,9 +61,25 @@ Task Master can provide a complexity report which can be helpful to read before
 Can you analyze the complexity of our tasks to help me understand which ones need to be broken down further?
 ```
 
+The agent will use the `analyze_project_complexity` MCP tool, or you can run it directly with the CLI command:
+```bash
+task-master analyze-complexity
+```
+
+For more comprehensive analysis using your configured research model, you can use:
+```bash
+task-master analyze-complexity --research
+```
+
+<Tip>
+The `--research` flag uses whatever research model you have configured in `.taskmaster/config.json` (configurable via `task-master models --setup`) for research-backed complexity analysis, providing more informed recommendations.
+</Tip>
+
 You can view the report in a friendly table using:
 ```
 Can you show me the complexity report in a more readable format?
 ```
 
-<Check>Now you are ready to begin [executing tasks](/docs/getting-started/quick-start/execute-quick)</Check>
+For more detailed CLI options, see the [Analyze Task Complexity](/capabilities/cli-root-commands#analyze-task-complexity) section.
+
+<Check>Now you are ready to begin [executing tasks](/getting-started/quick-start/execute-quick)</Check>

apps/docs/introduction.mdx

@@ -4,7 +4,7 @@ Welcome to v1 of the Task Master Docs. Expect weekly updates as we expand and re
 
 We've organized the docs into three sections depending on your experience level and goals:
 
-### Getting Started - Jump in to [Quick Start](/docs/getting-started/quick-start)
+### Getting Started - Jump in to [Quick Start](/getting-started/quick-start)
 Designed for first-time users. Get set up, create your first PRD, and run your first task.
 
 ### Best Practices

apps/docs/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "docs",
-	"version": "0.0.2",
+	"version": "0.0.4",
 	"private": true,
 	"description": "Task Master documentation powered by Mintlify",
 	"scripts": {

apps/extension/CHANGELOG.md

@@ -1,5 +1,50 @@
 # Change Log
 
+## 0.25.3
+
+### Patch Changes
+
+- Updated dependencies [[`044a7bf`](https://github.com/eyaltoledano/claude-task-master/commit/044a7bfc98049298177bc655cf341d7a8b6a0011)]:
+  - task-master-ai@0.27.2
+
+## 0.25.2
+
+### Patch Changes
+
+- Updated dependencies [[`f487736`](https://github.com/eyaltoledano/claude-task-master/commit/f487736670ef8c484059f676293777eabb249c9e), [`c911608`](https://github.com/eyaltoledano/claude-task-master/commit/c911608f60454253f4e024b57ca84e5a5a53f65c), [`1a18794`](https://github.com/eyaltoledano/claude-task-master/commit/1a1879483b86c118a4e46c02cbf4acebfcf6bcf9)]:
+  - task-master-ai@0.27.1
+
+## 0.25.2-rc.1
+
+### Patch Changes
+
+- Updated dependencies [[`1a18794`](https://github.com/eyaltoledano/claude-task-master/commit/1a1879483b86c118a4e46c02cbf4acebfcf6bcf9)]:
+  - task-master-ai@0.27.1-rc.1
+
+## 0.25.2-rc.0
+
+### Patch Changes
+
+- Updated dependencies [[`f487736`](https://github.com/eyaltoledano/claude-task-master/commit/f487736670ef8c484059f676293777eabb249c9e)]:
+  - task-master-ai@0.27.1-rc.0
+
+## 0.25.0
+
+### Minor Changes
+
+- [#1200](https://github.com/eyaltoledano/claude-task-master/pull/1200) [`fce8414`](https://github.com/eyaltoledano/claude-task-master/commit/fce841490a9ebbf1801a42dd8a29397379cf1142) Thanks [@eyaltoledano](https://github.com/eyaltoledano)! - Add "Start Task" button to VS Code extension for seamless Claude Code integration
+
+  You can now click a "Start Task" button directly in the Task Master extension which will open a new terminal and automatically execute the task using Claude Code. This provides a seamless workflow from viewing tasks in the extension to implementing them without leaving VS Code.
+
+- [#1201](https://github.com/eyaltoledano/claude-task-master/pull/1201) [`83af314`](https://github.com/eyaltoledano/claude-task-master/commit/83af314879fc0e563581161c60d2bd089899313e) Thanks [@losolosol](https://github.com/losolosol)! - Added a Start Build button to the VSCODE Task Properties Right Panel
+
+### Patch Changes
+
+- [#1229](https://github.com/eyaltoledano/claude-task-master/pull/1229) [`674d1f6`](https://github.com/eyaltoledano/claude-task-master/commit/674d1f6de7ea98116b61bdae6198bafe6c4e7c1a) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Fix MCP not connecting to new Taskmaster version
+
+- Updated dependencies [[`4e12643`](https://github.com/eyaltoledano/claude-task-master/commit/4e126430a092fb54afb035514fb3d46115714f97), [`fce8414`](https://github.com/eyaltoledano/claude-task-master/commit/fce841490a9ebbf1801a42dd8a29397379cf1142), [`fce8414`](https://github.com/eyaltoledano/claude-task-master/commit/fce841490a9ebbf1801a42dd8a29397379cf1142), [`fce8414`](https://github.com/eyaltoledano/claude-task-master/commit/fce841490a9ebbf1801a42dd8a29397379cf1142), [`a621ff0`](https://github.com/eyaltoledano/claude-task-master/commit/a621ff05eafb51a147a9aabd7b37ddc0e45b0869), [`e6de285`](https://github.com/eyaltoledano/claude-task-master/commit/e6de285ceacb0a397e952a63435cd32a9c731515), [`fce8414`](https://github.com/eyaltoledano/claude-task-master/commit/fce841490a9ebbf1801a42dd8a29397379cf1142)]:
+  - task-master-ai@0.27.0
+
 ## 0.25.0-rc.0
 
 ### Minor Changes

apps/extension/README.md

@@ -109,7 +109,7 @@ Access settings via **File → Preferences → Settings** and search for "Taskma
 
 ### **MCP Connection Settings**
 - **MCP Server Command** - Path to task-master-ai executable (default: `npx`)
-- **MCP Server Args** - Arguments for the server command (default: `-y`, `--package=task-master-ai`, `task-master-ai`)
+- **MCP Server Args** - Arguments for the server command (default: `-y`, `task-master-ai`)
 - **Connection Timeout** - Server response timeout (default: 30s)
 - **Auto Refresh** - Enable automatic task updates (default: enabled)
 

apps/extension/package.json

@@ -3,7 +3,7 @@
 	"private": true,
 	"displayName": "TaskMaster",
 	"description": "A visual Kanban board interface for TaskMaster projects in VS Code",
-	"version": "0.25.0-rc.0",
+	"version": "0.25.3",
 	"publisher": "Hamster",
 	"icon": "assets/icon.png",
 	"engines": {
@@ -240,7 +240,7 @@
 		"check-types": "tsc --noEmit"
 	},
 	"dependencies": {
-		"task-master-ai": "*"
+		"task-master-ai": "0.27.2"
 	},
 	"devDependencies": {
 		"@dnd-kit/core": "^6.3.1",
@@ -254,8 +254,9 @@
 		"@radix-ui/react-separator": "^1.1.7",
 		"@radix-ui/react-slot": "^1.2.3",
 		"@tailwindcss/postcss": "^4.1.11",
+		"@tanstack/react-query": "^5.83.0",
 		"@types/mocha": "^10.0.10",
-		"@types/node": "20.x",
+		"@types/node": "^22.10.5",
 		"@types/react": "19.1.8",
 		"@types/react-dom": "19.1.6",
 		"@types/vscode": "^1.101.0",
@@ -271,12 +272,11 @@
 		"lucide-react": "^0.525.0",
 		"npm-run-all": "^4.1.5",
 		"postcss": "8.5.6",
+		"react": "^19.0.0",
+		"react-dom": "^19.0.0",
 		"tailwind-merge": "^3.3.1",
 		"tailwindcss": "4.1.11",
-		"typescript": "^5.8.3",
-		"@tanstack/react-query": "^5.83.0",
-		"react": "^19.0.0",
-		"react-dom": "^19.0.0"
+		"typescript": "^5.7.3"
 	},
 	"overrides": {
 		"glob@<8": "^10.4.5",

apps/extension/package.publish.json

@@ -95,7 +95,7 @@
 					"items": {
 						"type": "string"
 					},
-					"default": ["-y", "--package=task-master-ai", "task-master-ai"],
+					"default": ["-y", "task-master-ai"],
 					"description": "An array of arguments to pass to the MCP server command."
 				},
 				"taskmaster.mcp.cwd": {

apps/extension/src/components/TaskDetails/TaskMetadataSidebar.tsx

@@ -11,7 +11,6 @@ interface TaskMetadataSidebarProps {
 	tasks: TaskMasterTask[];
 	complexity: any;
 	isSubtask: boolean;
-	sendMessage: (message: any) => Promise<any>;
 	onStatusChange: (status: TaskMasterTask['status']) => void;
 	onDependencyClick: (depId: string) => void;
 	isRegenerating?: boolean;
@@ -23,13 +22,12 @@ export const TaskMetadataSidebar: React.FC<TaskMetadataSidebarProps> = ({
 	tasks,
 	complexity,
 	isSubtask,
-	sendMessage,
 	onStatusChange,
 	onDependencyClick,
 	isRegenerating = false,
 	isAppending = false
 }) => {
-	const { vscode } = useVSCodeContext();
+	const { sendMessage } = useVSCodeContext();
 	const [isLoadingComplexity, setIsLoadingComplexity] = useState(false);
 	const [mcpComplexityScore, setMcpComplexityScore] = useState<
 		number | undefined
@@ -101,26 +99,37 @@ export const TaskMetadataSidebar: React.FC<TaskMetadataSidebarProps> = ({
 	};
 
 	// Handle starting a task
-	const handleStartTask = () => {
+	const handleStartTask = async () => {
 		if (!currentTask || isStartingTask) {
 			return;
 		}
 
 		setIsStartingTask(true);
 
-		// Send message to extension to open terminal
-		if (vscode) {
-			vscode.postMessage({
+		try {
+			// Send message to extension to open terminal
+			const result = await sendMessage({
 				type: 'openTerminal',
-				taskId: currentTask.id,
-				taskTitle: currentTask.title
+				data: {
+					taskId: currentTask.id,
+					taskTitle: currentTask.title
+				}
 			});
-		}
 
-		// Reset loading state after a short delay
-		setTimeout(() => {
+			// Handle the response
+			if (result && !result.success) {
+				console.error('Terminal execution failed:', result.error);
+				// The extension will show VS Code error notification and webview toast
+			} else if (result && result.success) {
+				console.log('Terminal started successfully:', result.terminalName);
+			}
+		} catch (error) {
+			console.error('Failed to start task:', error);
+			// This handles network/communication errors
+		} finally {
+			// Reset loading state
 			setIsStartingTask(false);
-		}, 500);
+		}
 	};
 
 	// Effect to handle complexity on task change

apps/extension/src/components/TaskDetailsView.tsx

@@ -208,7 +208,6 @@ export const TaskDetailsView: React.FC<TaskDetailsViewProps> = ({
 					tasks={allTasks}
 					complexity={complexity}
 					isSubtask={isSubtask}
-					sendMessage={sendMessage}
 					onStatusChange={handleStatusChange}
 					onDependencyClick={handleDependencyClick}
 				/>

apps/extension/src/extension.ts

@@ -8,6 +8,7 @@ import { ConfigService } from './services/config-service';
 import { PollingService } from './services/polling-service';
 import { createPollingStrategy } from './services/polling-strategies';
 import { TaskRepository } from './services/task-repository';
+import { TerminalManager } from './services/terminal-manager';
 import { WebviewManager } from './services/webview-manager';
 import { EventEmitter } from './utils/event-emitter';
 import { ExtensionLogger } from './utils/logger';
@@ -22,6 +23,7 @@ let logger: ExtensionLogger;
 let mcpClient: MCPClientManager;
 let api: TaskMasterApi;
 let repository: TaskRepository;
+let terminalManager: TerminalManager;
 let pollingService: PollingService;
 let webviewManager: WebviewManager;
 let events: EventEmitter;
@@ -46,6 +48,9 @@ export async function activate(context: vscode.ExtensionContext) {
 		// Repository with caching (actually useful for performance)
 		repository = new TaskRepository(api, logger);
 
+		// Terminal manager for task execution
+		terminalManager = new TerminalManager(context, logger);
+
 		// Config service for TaskMaster config.json
 		configService = new ConfigService(logger);
 
@@ -56,7 +61,13 @@ export async function activate(context: vscode.ExtensionContext) {
 		pollingService = new PollingService(repository, strategy, logger);
 
 		// Webview manager (cleaner than global panel array) - create before connection
-		webviewManager = new WebviewManager(context, repository, events, logger);
+		webviewManager = new WebviewManager(
+			context,
+			repository,
+			events,
+			logger,
+			terminalManager
+		);
 		webviewManager.setConfigService(configService);
 
 		// Sidebar webview manager
@@ -210,10 +221,11 @@ function registerCommands(context: vscode.ExtensionContext) {
 	);
 }
 
-export function deactivate() {
+export async function deactivate() {
 	logger?.log('👋 TaskMaster Extension deactivating...');
 	pollingService?.stop();
 	webviewManager?.dispose();
+	await terminalManager?.dispose();
 	api?.destroy();
 	mcpClient?.disconnect();
 }

apps/extension/src/services/terminal-manager.ts

@@ -0,0 +1,156 @@
+/**
+ * Terminal Manager - Handles task execution in VS Code terminals
+ * Uses @tm/core for consistent task management with the CLI
+ */
+
+import * as vscode from 'vscode';
+import { createTaskMasterCore, type TaskMasterCore } from '@tm/core';
+import type { ExtensionLogger } from '../utils/logger';
+
+export interface TerminalExecutionOptions {
+	taskId: string;
+	taskTitle: string;
+	tag?: string;
+}
+
+export interface TerminalExecutionResult {
+	success: boolean;
+	error?: string;
+	terminalName?: string;
+}
+
+export class TerminalManager {
+	private terminals = new Map<string, vscode.Terminal>();
+	private tmCore?: TaskMasterCore;
+
+	constructor(
+		private context: vscode.ExtensionContext,
+		private logger: ExtensionLogger
+	) {}
+
+	/**
+	 * Execute a task in a new VS Code terminal with Claude
+	 * Uses @tm/core for consistent task management with the CLI
+	 */
+	async executeTask(
+		options: TerminalExecutionOptions
+	): Promise<TerminalExecutionResult> {
+		const { taskTitle, tag } = options;
+		// Ensure taskId is always a string
+		const taskId = String(options.taskId);
+
+		this.logger.log(
+			`Starting task execution for ${taskId}: ${taskTitle}${tag ? ` (tag: ${tag})` : ''}`
+		);
+		this.logger.log(`TaskId type: ${typeof taskId}, value: ${taskId}`);
+
+		try {
+			// Initialize tm-core if needed
+			await this.initializeCore();
+
+			// Use tm-core to start the task (same as CLI)
+			const startResult = await this.tmCore!.startTask(taskId, {
+				dryRun: false,
+				force: false,
+				updateStatus: true
+			});
+
+			if (!startResult.started || !startResult.executionOutput) {
+				throw new Error(
+					startResult.error || 'Failed to start task with tm-core'
+				);
+			}
+
+			// Create terminal with custom TaskMaster icon
+			const terminalName = `Task ${taskId}: ${taskTitle}`;
+			const terminal = this.createTerminal(terminalName);
+
+			// Store terminal reference for potential cleanup
+			this.terminals.set(taskId, terminal);
+
+			// Show terminal and run Claude command
+			terminal.show();
+			const command = `claude "${startResult.executionOutput}"`;
+			terminal.sendText(command);
+
+			this.logger.log(`Launched Claude for task ${taskId} using tm-core`);
+
+			return {
+				success: true,
+				terminalName
+			};
+		} catch (error) {
+			this.logger.error('Failed to execute task:', error);
+			return {
+				success: false,
+				error: error instanceof Error ? error.message : 'Unknown error'
+			};
+		}
+	}
+
+	/**
+	 * Create a new terminal with TaskMaster branding
+	 */
+	private createTerminal(name: string): vscode.Terminal {
+		const workspaceRoot = vscode.workspace.workspaceFolders?.[0]?.uri.fsPath;
+
+		return vscode.window.createTerminal({
+			name,
+			cwd: workspaceRoot,
+			iconPath: new vscode.ThemeIcon('play') // Use a VS Code built-in icon for now
+		});
+	}
+
+	/**
+	 * Initialize TaskMaster Core (same as CLI)
+	 */
+	private async initializeCore(): Promise<void> {
+		if (!this.tmCore) {
+			const workspaceRoot = vscode.workspace.workspaceFolders?.[0]?.uri.fsPath;
+			if (!workspaceRoot) {
+				throw new Error('No workspace folder found');
+			}
+			this.tmCore = await createTaskMasterCore({ projectPath: workspaceRoot });
+		}
+	}
+
+	/**
+	 * Get terminal by task ID (if still active)
+	 */
+	getTerminalByTaskId(taskId: string): vscode.Terminal | undefined {
+		return this.terminals.get(taskId);
+	}
+
+	/**
+	 * Clean up terminated terminals
+	 */
+	cleanupTerminal(taskId: string): void {
+		const terminal = this.terminals.get(taskId);
+		if (terminal) {
+			this.terminals.delete(taskId);
+		}
+	}
+
+	/**
+	 * Dispose all managed terminals and clean up tm-core
+	 */
+	async dispose(): Promise<void> {
+		this.terminals.forEach((terminal) => {
+			try {
+				terminal.dispose();
+			} catch (error) {
+				this.logger.error('Failed to dispose terminal:', error);
+			}
+		});
+		this.terminals.clear();
+
+		if (this.tmCore) {
+			try {
+				await this.tmCore.close();
+				this.tmCore = undefined;
+			} catch (error) {
+				this.logger.error('Failed to close tm-core:', error);
+			}
+		}
+	}
+}

apps/extension/src/services/webview-manager.ts

@@ -8,6 +8,7 @@ import type { EventEmitter } from '../utils/event-emitter';
 import type { ExtensionLogger } from '../utils/logger';
 import type { ConfigService } from './config-service';
 import type { TaskRepository } from './task-repository';
+import type { TerminalManager } from './terminal-manager';
 
 export class WebviewManager {
 	private panels = new Set<vscode.WebviewPanel>();
@@ -19,7 +20,8 @@ export class WebviewManager {
 		private context: vscode.ExtensionContext,
 		private repository: TaskRepository,
 		private events: EventEmitter,
-		private logger: ExtensionLogger
+		private logger: ExtensionLogger,
+		private terminalManager: TerminalManager
 	) {}
 
 	setConfigService(configService: ConfigService): void {
@@ -362,27 +364,67 @@ export class WebviewManager {
 					return;
 
 				case 'openTerminal':
-					// Open VS Code terminal for task execution
+					// Delegate terminal execution to TerminalManager
+					const { taskId, taskTitle } = data.data || data; // Handle both nested and direct data
 					this.logger.log(
-						`Opening terminal for task ${data.taskId}: ${data.taskTitle}`
+						`Webview openTerminal - taskId: ${taskId} (type: ${typeof taskId}), taskTitle: ${taskTitle}`
 					);
 
-					try {
-						const terminal = vscode.window.createTerminal({
-							name: `Task ${data.taskId}: ${data.taskTitle}`,
-							cwd: vscode.workspace.workspaceFolders?.[0]?.uri.fsPath
-						});
-						terminal.show();
+					// Get current tag to ensure we're working in the right context
+					let currentTag = 'master'; // default fallback
+					if (this.mcpClient) {
+						try {
+							const tagsResult = await this.mcpClient.callTool('list_tags', {
+								projectRoot: vscode.workspace.workspaceFolders?.[0]?.uri.fsPath,
+								showMetadata: false
+							});
 
-						this.logger.log('Terminal created and shown successfully');
-						response = { success: true };
-					} catch (error) {
-						this.logger.error('Failed to create terminal:', error);
-						response = {
-							success: false,
-							error: error instanceof Error ? error.message : 'Unknown error'
-						};
+							let parsedData;
+							if (
+								tagsResult?.content &&
+								Array.isArray(tagsResult.content) &&
+								tagsResult.content[0]?.text
+							) {
+								try {
+									parsedData = JSON.parse(tagsResult.content[0].text);
+									if (parsedData?.data?.currentTag) {
+										currentTag = parsedData.data.currentTag;
+									}
+								} catch (e) {
+									this.logger.warn(
+										'Failed to parse tags response for terminal execution'
+									);
+								}
+							}
+						} catch (error) {
+							this.logger.warn(
+								'Failed to get current tag for terminal execution:',
+								error
+							);
+						}
 					}
+
+					const result = await this.terminalManager.executeTask({
+						taskId,
+						taskTitle,
+						tag: currentTag
+					});
+
+					response = result;
+
+					// Show user feedback AFTER sending the response (like the working "TaskMaster connected!" example)
+					setImmediate(() => {
+						if (result.success) {
+							// Success: Show info message
+							vscode.window.showInformationMessage(
+								`✅ Started Claude session for Task ${taskId}: ${taskTitle}`
+							);
+						} else {
+							// Error: Show VS Code native error notification only
+							const errorMsg = `Failed to start task: ${result.error}`;
+							vscode.window.showErrorMessage(errorMsg);
+						}
+					});
 					break;
 
 				default:

apps/extension/src/utils/mcpClient.ts

@@ -408,7 +408,7 @@ export function createMCPConfigFromSettings(): MCPConfig {
 			const taskMasterPath = require.resolve('task-master-ai');
 			const mcpServerPath = path.resolve(
 				path.dirname(taskMasterPath),
-				'mcp-server/server.js'
+				'./dist/mcp-server.js'
 			);
 
 			// Verify the server file exists

apps/extension/tsconfig.json

@@ -20,7 +20,8 @@
 		"paths": {
 			"@/*": ["./src/*"],
 			"@/components/*": ["./src/components/*"],
-			"@/lib/*": ["./src/lib/*"]
+			"@/lib/*": ["./src/lib/*"],
+			"@tm/core": ["../core/src"]
 		}
 	},
 	"exclude": ["node_modules", ".vscode-test", "out", "dist"]

assets/AGENTS.md

@@ -85,7 +85,7 @@ Task Master provides an MCP server that Claude Code can connect to. Configure in
   "mcpServers": {
     "task-master-ai": {
       "command": "npx",
-      "args": ["-y", "--package=task-master-ai", "task-master-ai"],
+      "args": ["-y", "task-master-ai"],
       "env": {
         "ANTHROPIC_API_KEY": "your_key_here",
         "PERPLEXITY_API_KEY": "your_key_here",