ai-flow-dev 1.3.0 → 1.4.0

package/prompts/backend/flow-dev-fix.md

@@ -229,7 +229,7 @@ Time: 4 minutes
 
 ```
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-📋 Fix Plan
+📋 Fix Plan (COMPLEX Bug)
 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 
 Issue: Memory leak in WebSocket connections
@@ -239,12 +239,85 @@ Root causes identified:
 2. Connection map not clearing disconnected sockets
 3. Redis subscriptions not being unsubscribed
 
-Fix plan:
-1. Add socket.removeAllListeners() on disconnect
-2. Delete from activeConnections Map on disconnect
-3. Call redisClient.unsubscribe() on disconnect
-4. Add heartbeat mechanism
-5. Add connection monitoring
+**Affected files:**
+- src/websocket/notificationSocket.ts (event listeners)
+- src/services/ConnectionManager.ts (connection map)
+- src/app.ts (Redis setup)
+
+**Estimated:** 5 SP (~8-10 hours) • 10 tasks
+
+**Fix plan:** (Test-First approach, execution order)
+
+- [ ] [T001] [P] Write test for event listener cleanup • 1 SP (~1-2h)
+      File: tests/unit/websocket/disconnect-cleanup.spec.ts
+      Tests: Verify listeners removed, memory not retained
+      Dependencies: None (can run parallel with T003)
+
+- [ ] [T002] Add socket.removeAllListeners() on disconnect • 1 SP (~1h)
+      File: src/websocket/notificationSocket.ts
+      Implements: Cleanup in disconnect event handler
+      Dependencies: None
+
+- [ ] [T003] [P] Write test for connection map cleanup • 1 SP (~1h)
+      File: tests/unit/services/ConnectionManager.spec.ts
+      Tests: Verify Map.delete() called, connection removed
+      Dependencies: None (can run parallel with T001)
+
+- [ ] [T004] Delete from activeConnections Map on disconnect • 1 SP (~30min)
+      File: src/services/ConnectionManager.ts
+      Implements: this.activeConnections.delete(socketId)
+      Dependencies: None
+
+- [ ] [T005] Write test for Redis unsubscribe • 1 SP (~1h)
+      File: tests/integration/redis/subscription-cleanup.spec.ts
+      Tests: Redis UNSUBSCRIBE called, channels cleaned up
+      Dependencies: None
+
+- [ ] [T006] Call redisClient.unsubscribe() on disconnect • 1 SP (~1h)
+      File: src/app.ts
+      Implements: Unsubscribe from all channels for this connection
+      Dependencies: None
+
+- [ ] [T007] Write test for heartbeat mechanism • 1 SP (~1-2h)
+      File: tests/unit/websocket/heartbeat.spec.ts
+      Tests: Ping/pong, timeout detection, stale cleanup
+      Dependencies: None
+
+- [ ] [T008] Implement heartbeat mechanism • 2 SP (~3-4h)
+      File: src/websocket/heartbeat.ts
+      Implements: Ping every 30s, timeout after 60s, auto-disconnect
+      Dependencies: None
+
+- [ ] [T009] Add connection monitoring and metrics • 1 SP (~1-2h)
+      File: src/services/ConnectionMonitor.ts
+      Implements: Track active count, memory usage, alert on leaks
+      Dependencies: T004 (needs ConnectionManager cleanup)
+
+- [ ] [T010] Integration test for complete cleanup flow • 1 SP (~1-2h)
+      File: tests/integration/websocket/full-cleanup.spec.ts
+      Tests: Connect 1000 clients, disconnect all, verify memory freed
+      Dependencies: T002, T004, T006, T008 (needs all fixes)
+
+**Parallelization Notes:**
+- T001, T003 can run in parallel (different test files)
+- T002, T004, T006, T008 can run in parallel (different files)
+- T010 (E2E test) depends on all fixes complete
+
+**Task Execution Graph:**
+
+```
+T001 [P] ──┐
+T003 [P] ──┼──> (Test tasks can run parallel)
+T005 [P] ──┤
+T007 [P] ──┘
+
+T002 (notificationSocket.ts) ──┐
+T004 (ConnectionManager.ts) ───┼──> T009 (Monitor) ──> T010 (E2E test)
+T006 (app.ts) ──────────────────┤
+T008 (heartbeat.ts) ────────────┘
+```
+
+**Estimated time:** 8-10 hours (with AI) | 2-3 days (manual)
 
 Proceed with fix? (Y/n)
 ```
@@ -900,4 +973,4 @@ Time: 14 minutes
 ---
 
 **BEGIN EXECUTION when user runs `/fix` or `/fix "description"`**
-
+

package/templates/{shared/AGENT.template.md → AGENT.template.md}

@@ -249,4 +249,4 @@ See `README.md` for installation and setup instructions.
 
 **Last Updated:** {{GENERATION_DATE}}
 
-**Generated by:** AI Flow v1.3.0
+**Generated by:** AI Flow v1.4.0

package/templates/mobile/AGENT.template.md

@@ -1,251 +0,0 @@
-# AGENT.md
-
-> Universal AI Assistant Configuration
->
-> This file provides context for ALL AI development tools (Claude, Copilot, Cursor, Gemini, etc.)
-
----
-
-## 📋 About This Project
-
-**Project Name:** {{PROJECT_NAME}}
-
-**Description:** {{PROJECT_DESCRIPTION}}
-
-**Problem We're Solving:** {{PROBLEM_STATEMENT}}
-
-**Target Users:** {{TARGET_USERS}}
-
-### Technical Context
-
-**Project Type:** Backend API/Service
-**Architecture:** {{ARCHITECTURE_PATTERN}}
-**Primary Language:** {{LANGUAGE}} {{LANGUAGE_VERSION}}
-**Framework:** {{FRAMEWORK}}
-**Database:** {{DATABASE}} with {{ORM}}
-
-**Key Characteristics:**
-- Authentication: {{AUTH_METHOD}}
-- API Style: {{API_STYLE}}
-- Deployment: {{DEPLOYMENT_PLATFORM}}
-- Current Phase: {{PROJECT_PHASE}}
-
-> This project uses AI-assisted development with comprehensive documentation.
-> All files below provide context to AI assistants for consistent, high-quality code generation.
-
----
-
-## 🏗️ Documentation Architecture
-
-This project follows **AI-assisted development** with comprehensive documentation.
-All documentation is structured to guide AI assistants in understanding the project deeply.
-
-### 📚 Core Documentation (Read in Order)
-
-1. **`ai-instructions.md`** ⭐ **START HERE**
-
-   - Tech stack and versions
-   - NEVER/ALWAYS rules
-   - Development workflow
-   - Priorities
-
-2. **`project-brief.md`**
-
-   - Business context
-   - Objectives and scope
-   - Constraints and success metrics
-
-3. **`README.md`**
-
-   - Quick start
-   - Installation and commands
-   - Project directory tour
-
-4. **`docs/architecture.md`**
-
-   - System architecture pattern
-   - Component structure
-   - Request/response flow
-
-5. **`docs/data-model.md`**
-
-   - Database schema
-   - Entity relationships
-   - Data patterns
-
-6. **`docs/code-standards.md`**
-
-   - Naming conventions
-   - Code quality rules
-   - Error handling patterns
-
-7. **`docs/testing.md`**
-
-   - Testing strategy
-   - Coverage requirements
-   - Test patterns
-
-8. **`docs/operations.md`**
-
-   - Deployment procedures
-   - Monitoring and alerts
-   - Scaling strategy
-
-9. **`docs/business-flows.md`**
-
-   - End-to-end business processes
-   - Key actors and systems
-   - Flow diagrams
-
-10. **`docs/api.md`**
-
-    - Endpoint conventions
-    - Authentication and pagination rules
-    - Request/response examples
-
-11. **`docs/contributing.md`**
-
-    - Development setup
-    - Contribution guidelines
-    - Code review process
-
-12. **`specs/security.md`**
-
-    - Authentication/Authorization
-    - Security policies
-    - Compliance requirements
-
-13. **`specs/configuration.md`**
-
-    - Environment variables
-    - External services
-    - Configuration by environment
-
-14. **`.env.example`**
-    - Required environment variables
-    - Default/local values
-    - Secrets handling guidance
-
----
-
-## ⚡ Quick Reference
-
-### Tech Stack
-
-**Backend:**
-
-- Framework: {{FRAMEWORK}}
-- Language: {{LANGUAGE}} {{LANGUAGE_VERSION}}
-- Database: {{DATABASE}}
-- ORM: {{ORM}}
-- Authentication: {{AUTH_METHOD}}
-- Caching: {{CACHE_STRATEGY}}
-
-**Infrastructure:**
-
-- Deployment: {{DEPLOYMENT_PLATFORM}}
-- CI/CD: {{CICD_PLATFORM}}
-- Monitoring: {{MONITORING_TOOLS}}
-
-### Critical Rules
-
-**❌ NEVER:**
-{{NEVER_RULES}}
-
-**✅ ALWAYS:**
-{{ALWAYS_RULES}}
-
----
-
-## 🤖 AI Assistant Workflow
-
-When working on this project:
-
-1. **Before starting ANY task:**
-
-   - Read `ai-instructions.md` for project-wide rules
-   - Check relevant documentation for the area you're working on
-   - Understand the business context from `project-brief.md`
-
-2. **When implementing features:**
-
-   - Follow architecture patterns from `docs/architecture.md`
-   - Respect data models from `docs/data-model.md`
-   - Apply code standards from `docs/code-standards.md`
-   - Add tests according to `docs/testing.md`
-
-3. **When handling security:**
-
-   - Consult `specs/security.md` for auth/authz
-   - Never hardcode secrets (use `specs/configuration.md`)
-   - Follow security headers and encryption requirements
-
-4. **When deploying:**
-   - Follow procedures in `docs/operations.md`
-   - Update configuration per environment
-   - Check health endpoints
-
----
-
-## 🛠️ Tool-Specific Configurations
-
-Different AI tools have specific configuration files that extend this AGENT.md:
-
-### Claude Code
-
-- **File:** `.clauderules`
-- **Purpose:** Claude-specific instructions and preferences
-- **References:** This AGENT.md + project docs
-
-### Cursor
-
-- **File:** `.cursorrules`
-- **Purpose:** Cursor-specific context and rules
-- **References:** This AGENT.md + project docs
-
-### GitHub Copilot
-
-- **File:** `.github/copilot-instructions.md`
-- **Purpose:** Copilot workspace instructions
-- **References:** This AGENT.md + project docs
-
-**All tool-specific configs reference this AGENT.md as the source of truth.**
-
----
-
-## 📊 Project Status
-
-**Current Phase:** {{PROJECT_PHASE}}
-
-**Architecture:** {{ARCHITECTURE_PATTERN}}
-
-**Compliance:** {{COMPLIANCE_REQUIREMENTS}}
-
----
-
-## 🚀 Getting Started
-
-### For AI Assistants
-
-1. Read this AGENT.md completely
-2. Read `ai-instructions.md` for critical rules
-3. Familiarize with `docs/architecture.md`
-4. Review `docs/code-standards.md` for coding patterns
-5. Check `specs/security.md` before handling auth/sensitive data
-
-### For Developers
-
-See `README.md` for installation and setup instructions.
-
----
-
-## 📝 Document Update Policy
-
-- Documents are **living artifacts** - update as project evolves
-- Breaking changes to architecture → Update `docs/architecture.md`
-- New security requirements → Update `specs/security.md`
-- Stack changes → Update `ai-instructions.md` AND this AGENT.md
-
-**Last Updated:** {{GENERATION_DATE}}
-
-**Generated by:** AI Flow v1.0.8