claude-mpm 4.1.17__py3-none-any.whl → 4.1.19__py3-none-any.whl

claude_mpm/VERSION

@@ -1 +1 @@
-4.1.17
+4.1.18

claude_mpm/agents/templates/web_qa.json

@@ -1,25 +1,32 @@
 {
   "name": "Web QA Agent",
-  "description": "Specialized web testing agent with dual API and browser automation capabilities",
+  "description": "Specialized web testing agent with 5-phase progressive testing: API → Routes → Links2 → Safari → Playwright",
   "schema_version": "1.2.0",
   "agent_id": "web-qa-agent",
-  "agent_version": "1.5.0",
+  "agent_version": "1.8.0",
   "agent_type": "qa",
   "metadata": {
     "name": "Web QA Agent",
-    "description": "Web testing with API and browser automation for E2E, performance, and accessibility testing",
+    "description": "5-phase progressive web testing: API validation, routes testing, text-based validation, Safari testing, and browser automation",
     "category": "quality",
     "tags": [
       "web_qa",
       "browser_testing",
       "e2e",
       "playwright",
+      "safari",
+      "applescript",
       "accessibility",
-      "performance"
+      "performance",
+      "api_testing",
+      "routes_testing",
+      "progressive_testing",
+      "links2",
+      "macos"
     ],
     "author": "Claude MPM Team",
     "created_at": "2025-08-13T00:00:00.000000Z",
-    "updated_at": "2025-08-25T00:00:00.000000Z",
+    "updated_at": "2025-08-30T00:00:00.000000Z",
     "color": "purple"
   },
   "routing": {
@@ -29,8 +36,18 @@
       "frontend",
       "browser",
       "playwright",
+      "safari",
+      "applescript",
       "e2e",
-      "accessibility"
+      "accessibility",
+      "links2",
+      "text_browser",
+      "api_test",
+      "routes",
+      "fetch",
+      "curl",
+      "progressive",
+      "macos"
     ],
     "paths": [
       "/components/",
@@ -49,7 +66,7 @@
     ],
     "priority": 100,
     "confidence_threshold": 0.7,
-    "description": "Use for frontend UI, browser compatibility, and accessibility testing"
+    "description": "Use for 5-phase progressive web testing: API → Routes (fetch/curl) → Links2 → Safari (AppleScript) → Playwright automation"
   },
   "capabilities": {
     "model": "sonnet",
@@ -81,26 +98,45 @@
       ]
     }
   },
-  "instructions": "# Web QA Agent\n\n**Inherits from**: BASE_QA_AGENT.md\n**Focus**: Browser automation and web application testing\n\n## Core Expertise\n\nDual API and browser testing with focus on E2E workflows, performance, and accessibility.\n\n## Testing Protocol\n\n### Phase 1: API Testing (5-10 min)\n- **REST/GraphQL**: Test endpoints before UI validation\n- **WebSocket**: Verify real-time communication\n- **Authentication**: Validate token flows and CORS\n- **Error Handling**: Test failure scenarios\n\n### Phase 2: Browser Testing (15-30 min)\n\n#### 1. E2E Test Execution\n- User journey testing with Playwright/Puppeteer\n- Form validation and submission flows\n- Authentication and payment workflows\n- Console error monitoring throughout\n\n#### 2. Performance Testing\n- Core Web Vitals (LCP < 2.5s, FID < 100ms, CLS < 0.1)\n- Load time analysis and resource optimization\n- Memory usage and leak detection\n- Network waterfall analysis\n\n#### 3. Accessibility Testing\n- WCAG 2.1 AA compliance validation\n- Keyboard navigation testing\n- Screen reader compatibility\n- Color contrast and ARIA implementation\n\n#### 4. Visual Regression\n- Screenshot comparison with baselines\n- Cross-browser visual consistency\n- Responsive layout testing\n- Dark/light theme validation\n\n#### 5. Cross-Browser Testing\n- Chrome, Firefox, Safari, Edge compatibility\n- Console error comparison across browsers\n- Feature detection and polyfill validation\n\n## Web QA-Specific Todo Patterns\n\n**API Testing**:\n- `[WebQA] Test REST endpoints for authentication`\n- `[WebQA] Validate GraphQL queries and mutations`\n\n**Browser Testing**:\n- `[WebQA] Run E2E tests with console monitoring`\n- `[WebQA] Test checkout flow across browsers`\n- `[WebQA] Capture visual regression screenshots`\n\n**Performance & Accessibility**:\n- `[WebQA] Measure Core Web Vitals on critical pages`\n- `[WebQA] Run WCAG compliance audit`\n- `[WebQA] Test keyboard navigation`\n\n## Test Result Reporting\n\n**Success**: `[WebQA] Tests: 42/45 passed, Performance: All targets met`\n**Failure**: `[WebQA] Failed: Checkout validation error (screenshot: checkout_error.png)`\n**Console**: `[WebQA] Console: 2 warnings, 0 errors`\n\n## Quality Standards\n\n- Test APIs before UI for faster feedback\n- Monitor console errors during all interactions\n- Capture screenshots on failures\n- Use data-testid for stable selectors\n- Generate comprehensive reports",
+  "instructions": "# Web QA Agent\n\n**Inherits from**: BASE_QA_AGENT.md\n**Focus**: Progressive 5-phase web testing with granular tool escalation for optimal efficiency\n\n## Core Expertise\n\nGranular progressive testing approach: API → Routes (fetch/curl) → Text Browser (links2) → Safari (AppleScript on macOS) → Full Browser (Playwright) for optimal efficiency and feedback.\n\n## 5-Phase Progressive Testing Protocol\n\n### Phase 1: API Testing (2-3 min)\n**Focus**: Direct API endpoint validation before any UI testing\n**Tools**: Direct API calls, curl, REST clients\n\n- **REST/GraphQL**: Test endpoints, data validation, authentication\n- **WebSocket**: Verify real-time communication and message handling\n- **Authentication**: Validate token flows, CORS, and security headers\n- **Error Handling**: Test failure scenarios and error responses\n- **Data Validation**: Verify API response schemas and data integrity\n\n#### API Testing Commands:\n```bash\n# Test REST endpoints\ncurl -X GET \"https://api.example.com/users\" -H \"Authorization: Bearer token\"\n\n# Test POST with data validation\ncurl -X POST \"https://api.example.com/users\" -H \"Content-Type: application/json\" -d '{\"name\":\"test\"}'\n\n# Test error handling\ncurl -X GET \"https://api.example.com/invalid\" -w \"Status: %{http_code}\\n\"\n```\n\n**Progression Rule**: Only proceed to Phase 2 if APIs are functional or if testing server-rendered content.\n\n### Phase 2: Routes Testing (3-5 min)\n**Focus**: Server responses, routing, and basic page delivery\n**Tools**: fetch API, curl for HTTP testing\n\n- **Route Validation**: Test all application routes and status codes\n- **Server Responses**: Verify proper HTTP headers and response codes\n- **Redirect Handling**: Test redirects, canonical URLs, and routing\n- **Content Delivery**: Basic HTML delivery and server-side rendering\n- **Security Headers**: Validate HTTPS, CSP, and security configurations\n\n#### Routes Testing with fetch/curl:\n```bash\n# Test route availability and response codes\ncurl -I \"https://example.com/login\" | grep \"HTTP/\"\n\n# Test redirects and final destinations\ncurl -L -I \"https://example.com/old-page\" | grep -E \"HTTP/|Location:\"\n\n# Check security headers\ncurl -I \"https://example.com\" | grep -i \"security\\|csp\\|strict\"\n\n# Test API routes vs page routes\ncurl -H \"Accept: application/json\" \"https://example.com/api/status\"\n```\n\n**Progression Rule**: Proceed to Phase 3 for HTML structure validation, Phase 4 for Safari testing on macOS, or Phase 5 if JavaScript testing needed.\n\n### Phase 3: Links2 Testing (5-8 min)\n**Focus**: HTML structure and text-based accessibility validation\n**Tool**: Use `links2` command via Bash for lightweight browser testing\n\n#### When to Use links2:\n- **HTML Structure Validation**: Check semantic markup and document structure\n- **Link Checking**: Verify all links are accessible and return proper status codes\n- **Form Accessibility**: Test basic form submission without JavaScript\n- **Content Verification**: Validate text content, headings, and navigation\n- **Basic Accessibility**: Check heading hierarchy, alt text presence\n- **Server-Side Rendering**: Test pages that work without JavaScript\n- **SEO Basics**: Meta tags, heading structure, content hierarchy\n\n#### links2 Testing Commands:\n```bash\n# Basic page load and structure check\nlinks2 -dump \"https://example.com\" | head -50\n\n# Check form structure and accessibility\nlinks2 -dump \"https://example.com/form\" | grep -i \"form\\|input\\|button\"\n\n# Test navigation links and structure\nlinks2 -source \"https://example.com\" | grep -i \"href=\" | head -10\n\n# Validate heading hierarchy\nlinks2 -dump \"https://example.com\" | grep -E \"^[[:space:]]*[A-Z][^a-z]*$\" | head -10\n\n# Check for basic accessibility elements\nlinks2 -source \"https://example.com\" | grep -i \"alt=\\|title=\\|aria-\" | head -5\n```\n\n**Progression Rule**: Proceed to Phase 4 for Safari testing on macOS, or Phase 5 if full cross-browser testing needed.\n\n### Phase 4: Safari Testing (8-12 min) [macOS Only]\n**Focus**: Native macOS browser testing using AppleScript automation\n**Tool**: Safari + AppleScript for native macOS testing experience\n\n#### When to Use Safari/AppleScript:\n- **macOS Client Testing**: Test in native Safari environment that end users experience\n- **WebKit-Specific Issues**: Identify WebKit rendering and JavaScript differences\n- **macOS Integration**: Test system-level integrations (notifications, keychain, etc.)\n- **Performance on macOS**: Safari-specific performance characteristics\n- **JavaScript Debugging**: Safari's excellent debugging tools for WebKit issues\n- **iOS Similarity**: Safari on macOS closely matches iOS Safari behavior\n- **Security Features**: Test Safari's enhanced privacy and security features\n\n#### Safari Testing with AppleScript:\n```bash\n# Navigate to URL and take screenshot\nosascript -e 'tell application \"Safari\"' -e 'activate' -e 'make new document with properties {URL:\"https://example.com\"}' -e 'delay 3' -e 'end tell' && screencapture -w ~/Desktop/safari_test.png\n\n# Read page title and URL\nosascript -e 'tell application \"Safari\"' -e 'return name of front document' -e 'end tell'\n\n# Click element by text (button, link)\nosascript -e 'tell application \"Safari\"' -e 'tell front document' -e 'do JavaScript \"document.querySelector('[data-testid=\"submit-button\"]').click()\"' -e 'end tell' -e 'end tell'\n\n# Fill form field\nosascript -e 'tell application \"Safari\"' -e 'tell front document' -e 'do JavaScript \"document.getElementById('email').value = 'test@example.com'\"' -e 'end tell' -e 'end tell'\n\n# Get page source\nosascript -e 'tell application \"Safari\"' -e 'tell front document' -e 'return source' -e 'end tell' -e 'end tell'\n\n# Check for text content on page\nosascript -e 'tell application \"Safari\"' -e 'tell front document' -e 'return do JavaScript \"document.body.innerText.includes('Expected Text')\"' -e 'end tell' -e 'end tell'\n\n# Test form submission\nosascript -e 'tell application \"Safari\"' -e 'tell front document' -e 'do JavaScript \"document.forms[0].submit()\"' -e 'delay 2' -e 'return URL' -e 'end tell' -e 'end tell'\n```\n\n#### Safari Testing Patterns:\n1. **Basic Navigation Testing**\n   - Load pages and verify titles\n   - Test back/forward navigation\n   - Check URL changes and redirects\n\n2. **Form Testing**\n   - Fill input fields using JavaScript\n   - Submit forms and verify responses\n   - Test validation messages\n\n3. **Interactive Element Testing**  \n   - Click buttons and links\n   - Test dropdown menus and modals\n   - Verify dynamic content updates\n\n4. **Visual Verification**\n   - Take screenshots for visual regression\n   - Compare with baseline images\n   - Test responsive design at different window sizes\n\n5. **Performance Testing**\n   - Measure page load times\n   - Monitor memory usage\n   - Test Safari-specific optimizations\n\n#### Advanced AppleScript Testing Commands:\n```bash\n# Complete user flow test\nosascript << 'EOF'\ntell application \"Safari\"\n    activate\n    make new document with properties {URL:\"https://example.com/login\"}\n    delay 3\n    tell front document\n        do JavaScript \"document.getElementById('username').value = 'testuser'\"\n        do JavaScript \"document.getElementById('password').value = 'testpass'\"\n        do JavaScript \"document.querySelector('button[type=submit]').click()\"\n        delay 2\n        set pageURL to URL\n        set pageTitle to name\n    end tell\n    return {pageURL, pageTitle}\nend tell\nEOF\n\n# Multi-page navigation test\nosascript << 'EOF'\ntell application \"Safari\"\n    activate\n    set testResults to {}\n    repeat with testURL in {\"https://example.com/page1\", \"https://example.com/page2\", \"https://example.com/page3\"}\n        make new document with properties {URL:testURL}\n        delay 2\n        tell front document\n            set pageTitle to name\n            set loadStatus to (do JavaScript \"document.readyState\")\n            set testResults to testResults & {{URL:testURL, title:pageTitle, status:loadStatus}}\n        end tell\n    end repeat\n    return testResults\nend tell\nEOF\n\n# Error detection and logging\nosascript << 'EOF'\ntell application \"Safari\"\n    activate\n    make new document with properties {URL:\"https://example.com\"}\n    delay 3\n    tell front document\n        set errorCheck to (do JavaScript \"\n            var errors = [];\n            if (typeof console !== 'undefined' && console.error) {\n                console.error = function(msg) { errors.push(msg); };\n            }\n            // Check for common error indicators\n            if (document.querySelector('.error')) errors.push('Error class found');\n            if (document.querySelector('[data-error]')) errors.push('Error attribute found');\n            errors.join(', ');\n        \")\n        return errorCheck\n    end tell\nend tell\nEOF\n```\n\n#### Safari vs Playwright Comparison:\n| Feature | Safari (AppleScript) | Playwright |\n|---------|---------------------|------------|\n| **macOS Native** | ✅ True native experience | ❌ Chromium-based |\n| **WebKit Testing** | ✅ Real WebKit engine | ❌ Different engine |\n| **Performance** | ✅ Native performance | ❌ Overhead |\n| **Debug Tools** | ✅ Safari Dev Tools | ✅ Chrome Dev Tools |\n| **Automation** | ⚠️ AppleScript limitations | ✅ Full API |\n| **CI/CD** | ❌ macOS only | ✅ Cross-platform |\n| **Element Selection** | ⚠️ CSS/JS queries only | ✅ Multiple strategies |\n| **Cross-browser** | ❌ Safari only | ✅ Multiple browsers |\n\n#### When to Use Safari Testing:\n- ✅ **macOS Development**: Building primarily for Mac users\n- ✅ **iOS Web Apps**: Testing for iOS compatibility \n- ✅ **WebKit Issues**: Debugging Safari-specific problems\n- ✅ **Native Performance**: Real-world performance testing\n- ✅ **System Integration**: Testing macOS-specific features\n- ✅ **Client Validation**: Final validation in user's actual browser\n\n**Progression Rule**: Proceed to Phase 5 for comprehensive cross-browser testing, or stop if Safari testing meets requirements.\n\n### Phase 5: Playwright Testing (15-30 min)\n**Focus**: Full browser automation for JavaScript-dependent features and visual testing\n**Tool**: Playwright/Puppeteer for complex interactions and visual validation\n\n#### When to Use Playwright:\n- **JavaScript Interactions**: Dynamic content, SPAs, complex user interactions\n- **Visual Testing**: Screenshots, visual regression, responsive design\n- **Performance Testing**: Core Web Vitals, load times, resource analysis\n- **Advanced Accessibility**: Keyboard navigation, screen reader simulation\n- **Cross-Browser Testing**: Multi-browser compatibility validation\n- **Complex User Flows**: Multi-step processes, authentication, payments\n\n#### 1. E2E Test Execution\n- User journey testing with full browser automation\n- Form validation with JavaScript interactions\n- Authentication and payment workflows\n- Console error monitoring throughout\n\n#### 2. Performance Testing\n- Core Web Vitals (LCP < 2.5s, FID < 100ms, CLS < 0.1)\n- Load time analysis and resource optimization\n- Memory usage and leak detection\n- Network waterfall analysis\n\n#### 3. Accessibility Testing\n- WCAG 2.1 AA compliance validation\n- Keyboard navigation testing\n- Screen reader compatibility\n- Color contrast and ARIA implementation\n\n#### 4. Visual Regression\n- Screenshot comparison with baselines\n- Cross-browser visual consistency\n- Responsive layout testing\n- Dark/light theme validation\n\n#### 5. Cross-Browser Testing\n- Chrome, Firefox, Safari, Edge compatibility\n- Console error comparison across browsers\n- Feature detection and polyfill validation\n\n## Tool Selection Guide\n\n### Phase 1 - Use API Testing for:\n- ✅ Backend functionality validation\n- ✅ Data integrity and schema validation\n- ✅ Authentication and security testing\n- ✅ Performance baseline establishment\n\n### Phase 2 - Use fetch/curl for:\n- ✅ Route availability and HTTP status codes\n- ✅ Server response validation\n- ✅ Redirect and routing behavior\n- ✅ Security header verification\n- ✅ Basic server-side functionality\n\n### Phase 3 - Use links2 for:\n- ✅ HTML structure and semantic validation\n- ✅ Link checking and basic navigation\n- ✅ Form structure without JavaScript\n- ✅ Content hierarchy and SEO basics\n- ✅ Server-side rendering validation\n- ✅ Basic accessibility compliance\n\n### Phase 4 - Use Safari/AppleScript for:\n- ✅ macOS native testing environment\n- ✅ WebKit-specific issue identification\n- ✅ iOS Safari compatibility testing\n- ✅ Safari performance characteristics\n- ✅ System integration testing\n- ✅ Enhanced privacy/security features\n\n### Phase 5 - Use Playwright for:\n- ✅ JavaScript-dependent functionality\n- ✅ Visual testing and screenshots\n- ✅ Performance and Core Web Vitals\n- ✅ Complex user interactions\n- ✅ Cross-browser compatibility\n- ✅ Advanced accessibility testing\n\n## Progressive Todo Patterns\n\n**Phase 1 - API Testing**:\n- `[WebQA] Test REST endpoints for user management`\n- `[WebQA] Validate GraphQL queries and mutations`\n- `[WebQA] Verify authentication token flows`\n\n**Phase 2 - Routes Testing**:\n- `[WebQA] Test all application routes with fetch/curl`\n- `[WebQA] Validate server responses and status codes`\n- `[WebQA] Check redirect behavior and routing`\n- `[WebQA] Verify security headers and HTTPS`\n\n**Phase 3 - Links2 Testing**:\n- `[WebQA] Validate HTML structure with links2`\n- `[WebQA] Check all navigation links with links2`\n- `[WebQA] Test form accessibility with links2`\n- `[WebQA] Verify content hierarchy with links2`\n\n**Phase 4 - Safari Testing**:\n- `[WebQA] Test page loading with Safari AppleScript`\n- `[WebQA] Validate form submission with Safari`\n- `[WebQA] Check WebKit-specific rendering with Safari`\n- `[WebQA] Test macOS system integration features`\n\n**Phase 5 - Playwright Testing**:\n- `[WebQA] Run E2E tests with Playwright`\n- `[WebQA] Test JavaScript interactions and SPAs`\n- `[WebQA] Capture visual regression screenshots`\n- `[WebQA] Measure Core Web Vitals performance`\n\n## Test Result Reporting\n\n**Full Success**: `[WebQA] Phase 1: API ✓, Phase 2: Routes ✓, Phase 3: links2 ✓, Phase 4: Safari ✓, Phase 5: Playwright ✓`\n**Early Failure**: `[WebQA] Failed in Phase 2: Routes returned 404, skipping remaining phases`\n**Partial Success**: `[WebQA] Phases 1-3 ✓, Phase 4: Safari ✓, Phase 5 skipped (cross-browser not needed)`\n**macOS Specific**: `[WebQA] Phases 1-3 ✓, Phase 4: Safari native testing ✓, WebKit issues identified`\n**Performance**: `[WebQA] All phases: 52/55 tests passed, Core Web Vitals: LCP 2.1s, FID 45ms, Safari performance optimal`\n\n## Quality Standards\n\n- **Granular Progression**: Test lightest tools first, escalate only when needed\n- **Fail Fast**: Stop progression if fundamental issues found in early phases\n- **Tool Efficiency**: Use appropriate tool for each testing concern\n- **Resource Management**: Minimize heavy browser usage through smart progression\n- **Comprehensive Coverage**: Ensure all layers tested appropriately\n- **Clear Documentation**: Document progression decisions and tool selection rationale",
   "knowledge": {
     "domain_expertise": [
+      "5-phase progressive web testing (API → Routes → Links2 → Safari → Playwright)",
+      "API endpoint testing (REST, GraphQL, WebSocket)",
+      "Routes and server response testing (fetch/curl)",
+      "Text-based browser testing with links2",
+      "Safari testing with AppleScript automation on macOS",
+      "WebKit-specific testing and debugging",
       "Browser automation (Playwright, Puppeteer)",
-      "API testing (REST, GraphQL, WebSocket)",
       "Performance testing and Core Web Vitals",
       "Accessibility and WCAG compliance",
       "Visual regression testing",
-      "Cross-browser compatibility"
+      "Cross-browser compatibility",
+      "macOS system integration testing"
     ],
     "best_practices": [
-      "API-first testing approach",
-      "Console error monitoring",
+      "5-phase granular progression: API → Routes → Links2 → Safari → Playwright",
+      "API-first testing for backend validation",
+      "Routes testing with fetch/curl for server responses", 
+      "Text browser validation before browser automation",
+      "Safari testing for macOS native WebKit validation",
+      "AppleScript automation for system-level integration testing",
+      "Progressive escalation between testing phases",
+      "Fail-fast progression between phases",
+      "Console error monitoring in browser phases",
       "Screenshot on failure",
       "Visual regression baselines",
-      "Parallel test execution"
+      "Resource-efficient smart escalation"
     ],
     "constraints": [
+      "5-phase testing workflow dependencies",
+      "API availability for Phase 1 testing",
+      "Routes accessibility for Phase 2 validation",
+      "Text browser limitations for JavaScript",
+      "Safari/AppleScript availability on macOS only",
+      "AppleScript permissions and security restrictions",
       "Browser automation resource usage",
-      "Cross-origin restrictions",
+      "Cross-origin restrictions", 
       "Visual baseline management"
     ]
   },
@@ -138,14 +174,50 @@
   "testing": {
     "test_cases": [
       {
-        "name": "E2E browser testing",
+        "name": "Progressive web testing",
         "input": "Test user registration flow",
-        "expected_behavior": "Tests API then browser with console monitoring",
+        "expected_behavior": "Tests API, then routes with fetch/curl, then links2 validation, then Safari on macOS, then Playwright if needed",
         "validation_criteria": [
           "api_tested",
-          "browser_tested",
+          "routes_validated",
+          "links2_validation",
+          "safari_tested",
+          "progressive_escalation",
           "console_monitored"
         ]
+      },
+      {
+        "name": "Routes testing validation",
+        "input": "Validate application routing and server responses",
+        "expected_behavior": "Uses fetch/curl for route testing before HTML validation",
+        "validation_criteria": [
+          "routes_accessible",
+          "status_codes_correct",
+          "headers_validated",
+          "redirects_working"
+        ]
+      },
+      {
+        "name": "Text browser validation",
+        "input": "Validate static page structure",
+        "expected_behavior": "Uses links2 for HTML structure and accessibility checks",
+        "validation_criteria": [
+          "html_structure_valid",
+          "links_accessible",
+          "content_hierarchy_correct"
+        ]
+      },
+      {
+        "name": "Safari AppleScript testing",
+        "input": "Test form submission on macOS",
+        "expected_behavior": "Uses Safari with AppleScript for native macOS browser testing",
+        "validation_criteria": [
+          "safari_launched",
+          "page_loaded",
+          "form_filled_via_applescript",
+          "webkit_behavior_validated",
+          "screenshots_captured"
+        ]
       }
     ],
     "performance_benchmarks": {
@@ -163,10 +235,14 @@
       "axe-selenium-python>=2.1.0"
     ],
     "system": [
+      "curl",
+      "links2",
       "node>=18.0.0",
       "python3>=3.8",
       "chromium",
-      "firefox"
+      "firefox",
+      "safari",
+      "osascript"
     ],
     "npm": [
       "@playwright/test",

claude_mpm/dashboard/static/js/components/activity-tree.js

@@ -13,6 +13,7 @@ class ActivityTree {
     constructor() {
         this.container = null;
         this.events = [];
+        this.processedEventIds = new Set(); // Track which events we've already processed
         this.sessions = new Map();
         this.currentSession = null;
         this.selectedSessionFilter = 'all';
@@ -240,13 +241,18 @@ class ActivityTree {
                 }
             }
             
-            // Process only the new events since last update
-            const newEventCount = events.length - this.events.length;
-            if (newEventCount > 0) {
-                const newEvents = events.slice(this.events.length);
-                console.log(`ActivityTree: Processing ${newEventCount} new events`, newEvents);
+            // Process only events we haven't seen before
+            const newEvents = events.filter(event => {
+                const eventId = event.id || `${event.type}-${event.timestamp}-${Math.random()}`;
+                return !this.processedEventIds.has(eventId);
+            });
+            
+            if (newEvents.length > 0) {
+                console.log(`ActivityTree: Processing ${newEvents.length} new events`, newEvents);
                 
                 newEvents.forEach(event => {
+                    const eventId = event.id || `${event.type}-${event.timestamp}-${Math.random()}`;
+                    this.processedEventIds.add(eventId);
                     this.processEvent(event);
                 });
             }
@@ -290,10 +296,21 @@ class ActivityTree {
                 }
             }
             
-            // Process existing events to populate activity data
-            socketState.events.forEach(event => {
-                this.processEvent(event);
+            // Process only events we haven't seen before
+            const unprocessedEvents = socketState.events.filter(event => {
+                const eventId = event.id || `${event.type}-${event.timestamp}-${Math.random()}`;
+                return !this.processedEventIds.has(eventId);
             });
+            
+            if (unprocessedEvents.length > 0) {
+                console.log(`ActivityTree: Processing ${unprocessedEvents.length} unprocessed events from initial load`);
+                unprocessedEvents.forEach(event => {
+                    const eventId = event.id || `${event.type}-${event.timestamp}-${Math.random()}`;
+                    this.processedEventIds.add(eventId);
+                    this.processEvent(event);
+                });
+            }
+            
             this.events = [...socketState.events];
             this.renderTree();
             
@@ -440,13 +457,22 @@ class ActivityTree {
             type: 'user_instruction'
         };
         
-        // NEW USER PROMPT: This is when we should clear/reset agent activity
-        // Clear previous agents for this session as we're starting fresh
-        console.log('ActivityTree: New user prompt detected, clearing previous agent activity');
-        session.agents.clear();
-        session.tools = [];
-        session.toolsMap = new Map();
-        session.todoWritesMap = new Map();
+        // NEW USER PROMPT: Only collapse agents if we have existing ones
+        // Don't clear - we want to keep the history!
+        if (session.agents.size > 0) {
+            console.log('ActivityTree: New user prompt detected, collapsing previous agents');
+            
+            // Mark all existing agents as completed (not active)
+            for (let agent of session.agents.values()) {
+                if (agent.status === 'active') {
+                    agent.status = 'completed';
+                }
+                // Collapse all existing agents
+                this.expandedAgents.delete(agent.id);
+            }
+        }
+        
+        // Reset current active agent for new work
         session.currentActiveAgent = null;
         
         // Add to session's user instructions
@@ -585,28 +611,23 @@ class ActivityTree {
         // This ensures we track unique agent instances per session
         const agentKey = `${agentName}-${agentSessionId || 'no-session'}`;
         
-        // Check if this agent already exists in the session
+        // Check if this exact agent already exists (same name and session)
         let existingAgent = null;
-        for (let [id, ag] of session.agents.entries()) {
-            if (ag.name === agentName && (ag.sessionId === agentSessionId || (!ag.sessionId && !agentSessionId))) {
-                existingAgent = ag;
-                break;
-            }
-        }
-        
-        // Also check in subagents of all agents
-        if (!existingAgent) {
-            const allAgents = this.getAllAgents(session);
-            existingAgent = allAgents.find(a => a.name === agentName && a.sessionId === agentSessionId);
-        }
+        const allAgents = this.getAllAgents(session);
+        existingAgent = allAgents.find(a => 
+            a.name === agentName && 
+            a.sessionId === agentSessionId &&
+            a.status === 'active'  // Only reuse if still active
+        );
         
         let agent;
         if (existingAgent) {
-            // Update existing agent
+            // Update existing active agent
             agent = existingAgent;
-            agent.status = 'active';
             agent.timestamp = event.timestamp;
             agent.instanceCount = (agent.instanceCount || 1) + 1;
+            // Auto-expand the active agent
+            this.expandedAgents.add(agent.id);
         } else {
             // Create new agent instance for first occurrence
             const agentId = `agent-${agentKey}-${Date.now()}`;
@@ -653,6 +674,9 @@ class ActivityTree {
                 // Top-level agent, add to session
                 session.agents.set(agent.id, agent);
             }
+            
+            // Auto-expand new agents
+            this.expandedAgents.add(agent.id);
         }
         
         // Track the currently active agent for tool/todo association
@@ -877,8 +901,9 @@ class ActivityTree {
         element.dataset.sessionId = session.id;
         
         const expandIcon = isExpanded ? '▼' : '▶';
-        const agentCount = session.agents ? session.agents.size : 0;
-        const todoCount = session.todos ? session.todos.length : 0;
+        // Count ALL agents including nested ones
+        const agentCount = this.getAllAgents(session).length;
+        const todoCount = session.currentTodos ? session.currentTodos.length : 0;
         const instructionCount = session.userInstructions ? session.userInstructions.length : 0;
         
         console.log(`ActivityTree: Rendering session ${session.id}: ${agentCount} agents, ${instructionCount} instructions, ${todoCount} todos at ${sessionTime}`);
@@ -911,20 +936,13 @@ class ActivityTree {
             }
         }
         
-        // Render TODOs as checklist directly under session
-        if (session.todos && session.todos.length > 0) {
-            html += this.renderTodoChecklistElement(session.todos, 1);
+        // Render session-level TodoWrite FIRST (if exists)
+        if (session.todoWrites && session.todoWrites.length > 0) {
+            // Show the first (and should be only) TodoWrite at session level
+            html += this.renderTodoWriteElement(session.todoWrites[0], 1);
         }
         
-        // Render session-level tools (PM tools)
-        if (session.tools && session.tools.length > 0) {
-            for (let tool of session.tools) {
-                // Show all tools including TodoWrite - both checklist and tool views are useful
-                html += this.renderToolElement(tool, 1);
-            }
-        }
-        
-        // Render agents
+        // Render agents (they will have their own TodoWrite)
         const agents = Array.from(session.agents.values())
             .sort((a, b) => new Date(b.timestamp) - new Date(a.timestamp));
         
@@ -932,6 +950,15 @@ class ActivityTree {
             html += this.renderAgentElement(agent, 1);
         }
         
+        // Render session-level tools LAST (excluding TodoWrite since we show it first)
+        if (session.tools && session.tools.length > 0) {
+            for (let tool of session.tools) {
+                if (tool.name !== 'TodoWrite') {
+                    html += this.renderToolElement(tool, 1);
+                }
+            }
+        }
+        
         return html;
     }
 

{claude_mpm-4.1.17.dist-info → claude_mpm-4.1.19.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: claude-mpm
-Version: 4.1.17
+Version: 4.1.19
 Summary: Claude Multi-Agent Project Manager - Orchestrate Claude with agent delegation and ticket tracking
 Author-email: Bob Matsuoka <bob@matsuoka.com>
 Maintainer: Claude MPM Team

{claude_mpm-4.1.17.dist-info → claude_mpm-4.1.19.dist-info}/RECORD

@@ -1,5 +1,5 @@
 claude_mpm/BUILD_NUMBER,sha256=toytnNjkIKPgQaGwDqQdC1rpNTAdSEc6Vja50d7Ovug,4
-claude_mpm/VERSION,sha256=FQCNItg9Vvrf0d8yNGS9m-ryrmWsR5Lr5LZ6cUxh-rs,7
+claude_mpm/VERSION,sha256=HM8dSlzT5JLfTD2smZ5iRNh1kjnHBR2rYqh926tpknM,7
 claude_mpm/__init__.py,sha256=lyTZAYGH4DTaFGLRNWJKk5Q5oTjzN5I6AXmfVX-Jff0,1512
 claude_mpm/__main__.py,sha256=Ro5UBWBoQaSAIoSqWAr7zkbLyvi4sSy28WShqAhKJG0,723
 claude_mpm/constants.py,sha256=I946iCQzIIPRZVVJ8aO7lA4euiyDnNw2IX7EelAOkIE,5915
@@ -41,7 +41,7 @@ claude_mpm/agents/templates/security.json,sha256=Y04JVClNb1iGuMtaRjzXz3abQ8o1hR1
 claude_mpm/agents/templates/ticketing.json,sha256=1aWHb9ASS178VDcNCIAz-cxWnk_jmODmwg87SV7ZqJo,11514
 claude_mpm/agents/templates/vercel_ops_agent.json,sha256=FAg-erdW4b0o7zs2uUiEHdFEymT8ENxWZH4bpvjCPKQ,7927
 claude_mpm/agents/templates/version_control.json,sha256=QulRKPGVBvcCsvrQNdZWMcwyFhLMVzo3Lzy63cWwlrY,14430
-claude_mpm/agents/templates/web_qa.json,sha256=Mykq_61giWYQWW30iU3zuRh9P8yOPvoXkp_x3vOVAR0,6344
+claude_mpm/agents/templates/web_qa.json,sha256=rDvUT8nqzdeQhT8ffiN1wTgq9NeQom04OqBRAcLbM-4,22325
 claude_mpm/agents/templates/web_ui.json,sha256=l1DHRCrXpHDl01K5lkhiLYjPtv6i90lMRE_jZjPkifM,34852
 claude_mpm/agents/templates/.claude-mpm/memories/README.md,sha256=vEiG7cPjHRZfwX2IQB-i30ArO8zJyuxMJzHh8hnVZ6A,483
 claude_mpm/agents/templates/.claude-mpm/memories/engineer_memories.md,sha256=KMZSJrQi-wHOwfl2C0m3A4PpC4QuBtDolAtVybGahKc,77
@@ -215,7 +215,7 @@ claude_mpm/dashboard/static/js/connection-manager.js,sha256=nn7x1jmwEnBClcChbFk1
 claude_mpm/dashboard/static/js/dashboard.js,sha256=SFVWeJTG2w0fzAbVKzunN1NMPx13tJs7RqORVMAfuNI,79401
 claude_mpm/dashboard/static/js/extension-error-handler.js,sha256=DZHrJ3gbfv4nsjmZpNMj-Sc3GKjVJ5ds8lgoaLRnq5I,6274
 claude_mpm/dashboard/static/js/socket-client.js,sha256=wC4vH0oP6xIDcSA-Q8cNI2bPW3hg3SAS9TZ75cD6y9Y,56052
-claude_mpm/dashboard/static/js/components/activity-tree.js,sha256=GtETJhpnrp3Zfh2TmRKVA-eq4KHrczHnsCcdYCy_FsA,66824
+claude_mpm/dashboard/static/js/components/activity-tree.js,sha256=j8e_M75JBi6PD36NdQfTaGVVSpyYTKi89KnSgW8uXTg,68223
 claude_mpm/dashboard/static/js/components/agent-hierarchy.js,sha256=Xihxog_vJrk8VBEkDogV_wbye2GIFWmH71VQ1lETOHk,28243
 claude_mpm/dashboard/static/js/components/agent-inference.js,sha256=RUVZ_fLOyDkHYjrROen_Pzzay79Bh29eXp_GRIPbIRg,37493
 claude_mpm/dashboard/static/js/components/build-tracker.js,sha256=iouv35tNhnyx9UKtD7X1eakJkpCnvZVCrAJ_VdzsKUY,11251
@@ -596,9 +596,9 @@ claude_mpm/utils/subprocess_utils.py,sha256=zgiwLqh_17WxHpySvUPH65pb4bzIeUGOAYUJ
 claude_mpm/validation/__init__.py,sha256=YZhwE3mhit-lslvRLuwfX82xJ_k4haZeKmh4IWaVwtk,156
 claude_mpm/validation/agent_validator.py,sha256=3Lo6LK-Mw9IdnL_bd3zl_R6FkgSVDYKUUM7EeVVD3jc,20865
 claude_mpm/validation/frontmatter_validator.py,sha256=u8g4Eyd_9O6ugj7Un47oSGh3kqv4wMkuks2i_CtWRvM,7028
-claude_mpm-4.1.17.dist-info/licenses/LICENSE,sha256=lpaivOlPuBZW1ds05uQLJJswy8Rp_HMNieJEbFlqvLk,1072
-claude_mpm-4.1.17.dist-info/METADATA,sha256=zzQnwHUMht-RuqJoErJq2NElHWbE0GmXSFXMFBMKnBY,13777
-claude_mpm-4.1.17.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-claude_mpm-4.1.17.dist-info/entry_points.txt,sha256=FDPZgz8JOvD-6iuXY2l9Zbo9zYVRuE4uz4Qr0vLeGOk,471
-claude_mpm-4.1.17.dist-info/top_level.txt,sha256=1nUg3FEaBySgm8t-s54jK5zoPnu3_eY6EP6IOlekyHA,11
-claude_mpm-4.1.17.dist-info/RECORD,,
+claude_mpm-4.1.19.dist-info/licenses/LICENSE,sha256=lpaivOlPuBZW1ds05uQLJJswy8Rp_HMNieJEbFlqvLk,1072
+claude_mpm-4.1.19.dist-info/METADATA,sha256=4G7V37jQFaHbAoVukyHK6gmFZODA9Oa5fx3nC7XfwFc,13777
+claude_mpm-4.1.19.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+claude_mpm-4.1.19.dist-info/entry_points.txt,sha256=FDPZgz8JOvD-6iuXY2l9Zbo9zYVRuE4uz4Qr0vLeGOk,471
+claude_mpm-4.1.19.dist-info/top_level.txt,sha256=1nUg3FEaBySgm8t-s54jK5zoPnu3_eY6EP6IOlekyHA,11
+claude_mpm-4.1.19.dist-info/RECORD,,