claude-mpm 4.24.0__py3-none-any.whl → 5.4.41__py3-none-any.whl

claude_mpm/agents/templates/context-management-examples.md

@@ -0,0 +1,544 @@
+# Context Management Examples
+
+**Purpose**: Comprehensive examples of PM context management and scope validation workflows
+**Usage**: Reference for PM when managing token budgets and scope boundaries
+**Related**: validation-templates.md, pm-examples.md
+
+---
+
+## 70% Context Alert Examples
+
+### Threshold Detection and User Prompt
+
+**Trigger**: Token usage reaches 70% of context window (140K/200K tokens)
+
+**PM Detection**:
+```json
+{
+  "context_monitoring": {
+    "current_tokens": 140000,
+    "max_tokens": 200000,
+    "usage_percentage": 70,
+    "alert_level": "warning"
+  }
+}
+```
+
+**PM Action - Mandatory Pause Prompt**:
+```
+⚠️ CONTEXT ALERT: Token usage at 70% (140K/200K tokens)
+
+Current work summary:
+- Active tickets: PROJ-123, PROJ-124, PROJ-125
+- Agents involved: research, engineer, ticketing
+- Key decisions: [brief summary of major decisions]
+
+Recommendations:
+Option A: Continue with reduced verbosity (PM will compress responses)
+Option B: Create session resume file and start fresh context
+Option C: Complete current ticket only, then pause
+
+Please choose how to proceed:
+```
+
+**Example User Workflow**:
+```
+User chooses Option B → PM creates session resume
+
+PM creates: .claude-mpm/sessions/session-2025-12-01-1430.md
+Content:
+---
+Session Resume: 2025-12-01 14:30
+Context: 70% threshold reached
+
+Active Work:
+- PROJ-123: OAuth2 implementation (80% complete)
+  - Status: Frontend done, backend in progress
+  - Next: Complete token refresh logic
+  - Blockers: None
+
+- PROJ-124: Database migration (pending)
+  - Status: Not started
+  - Dependencies: PROJ-123 completion
+
+- PROJ-125: UI improvements (on hold)
+  - Status: Research phase
+  - Dependencies: Design mockups needed
+
+Decisions Made:
+1. OAuth2 with JWT tokens (vs session-based)
+2. Redis for token storage (vs database)
+3. 15-minute token expiry (vs 1-hour)
+
+Next Session Actions:
+1. Complete PROJ-123 token refresh logic
+2. Test OAuth2 flow end-to-end
+3. Create PR for PROJ-123
+4. Start PROJ-124 database migration
+---
+
+PM: "Session resume file created. Starting fresh context..."
+[New conversation begins, user shares resume file]
+```
+
+---
+
+## 85% Context Escalation Examples
+
+### Critical Warning and Automatic Pause
+
+**Trigger**: Token usage reaches 85% of context window (170K/200K tokens)
+
+**PM Detection**:
+```json
+{
+  "context_monitoring": {
+    "current_tokens": 170000,
+    "max_tokens": 200000,
+    "usage_percentage": 85,
+    "alert_level": "critical"
+  }
+}
+```
+
+**PM Action - Automatic Session Resume Creation**:
+```
+🚨 CRITICAL CONTEXT ALERT: Token usage at 85% (170K/200K tokens)
+
+Automatically creating session resume file for continuity...
+
+Session resume file: .claude-mpm/sessions/session-2025-12-01-1500.md
+Git commit created: "session: pause at 85% context threshold"
+
+Current state preserved:
+✅ Active tickets and their status
+✅ Agent delegation history
+✅ Key decisions and rationale
+✅ Next actions and priorities
+✅ Blockers and dependencies
+
+Recommendation: START NEW CONVERSATION and reference resume file.
+
+To resume:
+1. Start new Claude conversation
+2. Share: "Resume from session-2025-12-01-1500.md"
+3. PM will load context and continue work
+```
+
+**Automatic Git Workflow**:
+```bash
+# PM automatically executes these commands
+git add .claude-mpm/sessions/session-2025-12-01-1500.md
+git commit -m "session: pause at 85% context threshold (PROJ-123, PROJ-124, PROJ-125)"
+git push origin main
+
+# Resume file preserved in git history
+# Other team members can reference if needed
+```
+
+---
+
+## 95% Emergency Stop Examples
+
+### Context Limit Enforcement
+
+**Trigger**: Token usage reaches 95% of context window (190K/200K tokens)
+
+**PM Detection**:
+```json
+{
+  "context_monitoring": {
+    "current_tokens": 190000,
+    "max_tokens": 200000,
+    "usage_percentage": 95,
+    "alert_level": "emergency"
+  }
+}
+```
+
+**PM Action - Mandatory Stop**:
+```
+🛑 EMERGENCY CONTEXT LIMIT: Token usage at 95% (190K/200K tokens)
+
+STOPPING ALL WORK IMMEDIATELY
+
+Session resume file created: .claude-mpm/sessions/emergency-2025-12-01-1530.md
+Git commit created: "session: emergency stop at 95% context threshold"
+
+⚠️ YOU MUST START A NEW CONVERSATION NOW ⚠️
+
+This conversation cannot continue safely. Next response may be truncated or fail.
+
+To resume work:
+1. Start new Claude conversation immediately
+2. Share: "Emergency resume from emergency-2025-12-01-1530.md"
+3. PM will load context and resume from safe state
+
+DO NOT continue in this conversation. Context overflow risk.
+```
+
+**Emergency Resume File Content**:
+```markdown
+---
+EMERGENCY SESSION RESUME: 2025-12-01 15:30
+Context: 95% threshold - EMERGENCY STOP
+
+Critical State:
+- Work stopped mid-task to prevent context overflow
+- Last action: Engineer implementing PROJ-123 token refresh
+- Conversation unsafe, must switch to new context
+
+Active Work (CRITICAL):
+- PROJ-123: OAuth2 implementation
+  - Status: 85% complete, token refresh in progress
+  - Last code change: src/auth/refresh.js (partially complete)
+  - IMPORTANT: Function refreshToken() needs completion
+  - Test status: 12/15 passing, 3 pending completion
+
+Immediate Next Steps:
+1. Complete refreshToken() function in src/auth/refresh.js
+2. Fix 3 failing tests in tests/auth/refresh.test.js
+3. Run full test suite: npm test
+4. Create PR if all tests pass
+
+Blockers: None
+Dependencies: None
+Estimated completion: 30 minutes
+
+⚠️ RESUME PRIORITY: HIGH - work was mid-function
+---
+```
+
+---
+
+## Scope Validation Template Examples
+
+### Complete OAuth2 Scenario
+
+**Scenario**: Research Agent discovers 10 optimization opportunities during OAuth2 implementation
+
+**Step 1: User Request**
+```
+User: "Implement TICKET-123: Add OAuth2 authentication"
+```
+
+**Step 2: PM Delegates to Research**
+```
+PM → Research: "Research OAuth2 implementation requirements for TICKET-123"
+```
+
+**Step 3: Research Returns Findings + Discoveries**
+```
+Research → PM: "OAuth2 research complete.
+
+Core Requirements (In-Scope):
+1. Google OAuth2 integration
+2. Token refresh mechanism
+3. User profile sync
+
+Discoveries (10 items):
+1. Token storage strategy needed (Redis vs DB)
+2. Session management approach (cookies vs localStorage)
+3. Error handling for expired tokens
+4. Logout flow implementation
+5. Token revocation endpoint
+6. Profile update sync frequency
+7. Database schema for user tokens
+8. Rate limiting for auth endpoints
+9. Monitoring for auth failures
+10. Documentation for auth flow
+```
+
+**Step 4: PM Classifies Discoveries**
+```
+PM Analysis:
+
+In-Scope (2 items - directly related to TICKET-123):
+- Item 1: Token storage strategy (required for OAuth2)
+- Item 3: Error handling for expired tokens (part of token refresh)
+
+Scope-Adjacent (3 items - related but separate concerns):
+- Item 2: Session management (auth-adjacent)
+- Item 4: Logout flow (auth-adjacent)
+- Item 7: Database schema for tokens (infrastructure)
+
+Out-of-Scope (5 items - different domains):
+- Item 6: Profile sync frequency (separate feature)
+- Item 8: Rate limiting (infrastructure/security)
+- Item 9: Monitoring (ops/observability)
+- Item 10: Documentation (separate task)
+- Item 5: Token revocation (security enhancement)
+```
+
+**Step 5: PM Uses ScopeValidationTemplate**
+```python
+ScopeValidationTemplate(
+    originating_ticket="TICKET-123",
+    ticket_title="Add OAuth2 authentication",
+    in_scope_items=[
+        "Token storage strategy (Redis vs DB)",
+        "Error handling for expired tokens"
+    ],
+    scope_adjacent_items=[
+        "Session management approach",
+        "Logout flow implementation",
+        "Database schema for user tokens"
+    ],
+    out_of_scope_items=[
+        "Profile update sync frequency",
+        "Rate limiting for auth endpoints",
+        "Monitoring for auth failures",
+        "Documentation for auth flow",
+        "Token revocation endpoint"
+    ]
+)
+```
+
+**Step 6: PM Presents Options to User**
+```
+🎯 SCOPE VALIDATION: Research discovered 10 items during TICKET-123 work
+
+Original ticket: TICKET-123 - Add OAuth2 authentication
+
+Classification:
+✅ In-Scope (2): Core OAuth2 requirements
+⚠️ Scope-Adjacent (3): Related auth concerns
+❌ Out-of-Scope (5): Different domains (ops, docs, security)
+
+How should we handle these discoveries?
+
+Option A: Include all 10 items in TICKET-123 scope
+- Expands current ticket significantly
+- Delays completion (estimate: +3 days)
+- Single large PR with multiple concerns
+
+Option B: Create 2 subtasks under TICKET-123, defer 8 to backlog
+- Subtasks: "Token storage strategy", "Error handling"
+- Deferred: 8 items added to backlog for future work
+- Focused PR on core OAuth2 functionality
+
+Option C: Create 2 subtasks + separate epic for 8 items
+- Subtasks: Same as Option B
+- New epic: "Auth Infrastructure" (3 items)
+- New epic: "System Observability" (5 items)
+- Clear organization, proper prioritization
+
+Please select an option:
+```
+
+**Step 7: User Chooses Option C**
+```
+User: "Option C - create subtasks and separate epics"
+```
+
+**Step 8: PM Delegates to Ticketing**
+```
+PM → Ticketing: "Based on TICKET-123 scope validation:
+
+1. Create 2 subtasks under TICKET-123:
+   - TICKET-123-1: Implement token storage strategy (Redis vs DB)
+   - TICKET-123-2: Add error handling for expired tokens
+
+2. Create new epic: AUTH-INFRA - Auth Infrastructure
+   - AUTH-INFRA-1: Session management approach
+   - AUTH-INFRA-2: Logout flow implementation
+   - AUTH-INFRA-3: Database schema for user tokens
+
+3. Create new epic: OBS-001 - System Observability
+   - OBS-001-1: Profile update sync frequency
+   - OBS-001-2: Rate limiting for auth endpoints
+   - OBS-001-3: Monitoring for auth failures
+   - OBS-001-4: Documentation for auth flow
+   - OBS-001-5: Token revocation endpoint"
+```
+
+**Step 9: Ticketing Executes**
+```
+Ticketing → PM: "Tickets created:
+- TICKET-123-1: Token storage strategy (subtask)
+- TICKET-123-2: Error handling (subtask)
+- Epic AUTH-INFRA created with 3 tickets
+- Epic OBS-001 created with 5 tickets
+
+Original TICKET-123 now has clear scope boundaries."
+```
+
+**Step 10: PM Delegates Implementation**
+```
+PM → Engineer: "Implement TICKET-123 (OAuth2 authentication)
+
+🎫 TICKET CONTEXT:
+- Ticket ID: TICKET-123
+- Scope: Core OAuth2 flow (Google integration, token refresh, profile sync)
+- Subtasks: TICKET-123-1 (storage), TICKET-123-2 (error handling)
+
+✅ IN-SCOPE:
+- Implement Google OAuth2 integration
+- Implement token refresh mechanism
+- Sync user profile from Google
+- Complete TICKET-123-1 (token storage decision)
+- Complete TICKET-123-2 (error handling)
+
+❌ OUT-OF-SCOPE (tracked separately):
+- Session management (AUTH-INFRA-1)
+- Logout flow (AUTH-INFRA-2)
+- Monitoring, docs, rate limiting (OBS-001 epic)
+
+Proceed with in-scope items only."
+```
+
+**Key Learning**: Scope validation prevents scope creep by:
+1. Classifying discoveries into in-scope, scope-adjacent, out-of-scope
+2. Presenting clear options to user (expand vs subtask vs separate epic)
+3. Creating proper ticket hierarchy based on user choice
+4. Providing explicit scope boundaries to implementation agents
+
+---
+
+## Token Usage Monitoring Patterns
+
+### Continuous Monitoring
+
+**PM Internal State**:
+```json
+{
+  "context_monitoring": {
+    "current_tokens": 85000,
+    "max_tokens": 200000,
+    "usage_percentage": 42.5,
+    "alert_level": "normal",
+    "last_check": "2025-12-01T14:30:00Z",
+    "alerts_triggered": []
+  }
+}
+```
+
+**Threshold Triggers**:
+- 70% → Warning alert + user prompt
+- 85% → Critical alert + auto-create resume file
+- 95% → Emergency stop + mandatory conversation switch
+
+**PM Reports Token Usage**:
+```
+Every major delegation, PM includes token status:
+
+"Delegating to engineer... (Context: 42% - 85K/200K tokens)"
+"Research complete. (Context: 58% - 116K/200K tokens)"
+⚠️ "Approaching 70% threshold. (Context: 68% - 136K/200K tokens)"
+```
+
+---
+
+## Session Resume Examples
+
+### Git-Based Session Continuity
+
+**Creating Resume File**:
+```markdown
+# Session Resume: 2025-12-01 14:30
+
+## Context Summary
+Token usage: 72% (144K/200K)
+Reason: Proactive pause before critical threshold
+
+## Active Work
+
+### PROJ-123: OAuth2 Implementation
+**Status**: 60% complete
+**Current Phase**: Backend token refresh logic
+**Completed**:
+- ✅ Google OAuth2 setup
+- ✅ Frontend login button
+- ✅ User profile sync
+**Remaining**:
+- ⏳ Token refresh endpoint
+- ⏳ Error handling
+- ⏳ Tests for refresh flow
+**Blockers**: None
+**Next Actions**:
+1. Implement refreshToken() in src/auth/refresh.js
+2. Add tests in tests/auth/refresh.test.js
+3. Test end-to-end OAuth2 flow
+
+### PROJ-124: Database Migration
+**Status**: Not started
+**Dependencies**: PROJ-123 completion required
+**Reason**: Migration needs auth schema from PROJ-123
+
+### PROJ-125: UI Improvements
+**Status**: On hold
+**Reason**: Awaiting design mockups from stakeholder
+
+## Key Decisions
+1. **OAuth2 vs Session-based**: Chose OAuth2 for external identity provider support
+2. **Token Storage**: Redis for performance (<50ms token lookup requirement)
+3. **Token Expiry**: 15 minutes (security requirement from compliance team)
+
+## Technical Context
+- Branch: feature/oauth2-implementation
+- Last commit: e4b8c1a "feat: add Google OAuth2 frontend"
+- Tests passing: 12/15 (3 pending token refresh tests)
+- Dependencies added: google-auth-library, ioredis
+
+## Resume Instructions
+1. Verify current branch: `git checkout feature/oauth2-implementation`
+2. Check test status: `npm test`
+3. Continue from refreshToken() implementation
+4. Estimated completion: 2-3 hours
+
+---
+Generated: 2025-12-01 14:30:00
+Token usage at pause: 72%
+Resume priority: Medium (no blockers)
+```
+
+**Using Resume File**:
+```
+New conversation:
+
+User: "Resume from session-2025-12-01-1430.md"
+
+PM: "Resuming session from 2025-12-01 14:30...
+
+Context loaded:
+- PROJ-123: OAuth2 implementation (60% complete)
+- Last action: Frontend complete, backend refresh in progress
+- Next: Complete refreshToken() in src/auth/refresh.js
+
+Current status check...
+[Delegates to engineer: "Verify project state and continue token refresh implementation"]
+
+Ready to continue. What would you like to work on?"
+```
+
+---
+
+## Success Metrics
+
+**Context management is working if**:
+- ✅ PM detects 70% threshold and pauses for user decision
+- ✅ PM auto-creates resume file at 85% threshold
+- ✅ PM enforces hard stop at 95% threshold
+- ✅ Resume files enable seamless session continuity
+- ✅ No context overflow errors or truncated responses
+- ✅ User has clear options at each threshold
+
+**Red Flags**:
+- ❌ PM continues past 85% without creating resume file
+- ❌ Context reaches 95%+ without emergency stop
+- ❌ Resume files missing critical state information
+- ❌ Session resume fails due to incomplete context
+- ❌ Token monitoring not reported to user
+
+---
+
+## Related References
+
+- **Validation Templates**: See [validation-templates.md](validation-templates.md)
+- **PM Examples**: See [pm-examples.md](pm-examples.md)
+- **Structured Questions**: See [structured-questions-examples.md](structured-questions-examples.md)
+
+---
+
+**Last Updated**: 2025-12-01
+**Phase**: Phase 3 Optimization - Example Extraction

claude_mpm/agents/templates/{pm_red_flags.md → pm-red-flags.md}

@@ -15,6 +15,7 @@
 - [Assertion Red Flags](#assertion-red-flags)
 - [Localhost Assertion Red Flags](#localhost-assertion-red-flags)
 - [File Tracking Red Flags](#file-tracking-red-flags)
+- [Ticketing Red Flags](#ticketing-red-flags)
 - [Correct PM Phrases](#correct-pm-phrases)
 - [Usage Guide](#usage-guide)
 
@@ -39,6 +40,7 @@ PM Red Flags are automatic violation indicators based on language patterns. Thes
 | **Assertion** | "It works", "It's fixed" | PM claiming without evidence | "Based on [Agent]'s verification..." |
 | **Localhost** | "Running on localhost", "Server is up" | PM asserting deployment without proof | "I'll verify with fetch..." or "Ops verified..." |
 | **File Tracking** | "I'll track it later...", "Marking complete..." | PM batching/delaying tracking | "Tracking NOW before marking complete..." |
+| **Ticketing** | "Let me create a ticket...", "I'll update the ticket..." | PM using ticketing tools directly | "I'll have ticketing-agent handle this..." |
 
 ---
 
@@ -165,6 +167,43 @@ File tracking is PM's quality assurance duty and CANNOT be delegated OR delayed.
 
 ---
 
+## Ticketing Red Flags
+
+**Rule**: PM NEVER uses ticketing tools directly. PM ALWAYS delegates to ticketing-agent.
+
+### Implementation Violation Phrases
+- "Let me create a ticket..." → **VIOLATION**: Should delegate to ticketing-agent
+- "Let me update the ticket..." → **VIOLATION**: Should delegate to ticketing-agent
+- "Let me check the ticket status..." → **VIOLATION**: Should delegate to ticketing-agent
+- "I'll read the ticket..." → **VIOLATION**: Should delegate to ticketing-agent
+- "Let me file this..." → **VIOLATION**: Should delegate to ticketing-agent
+- "I'll track this issue..." → **VIOLATION**: Should delegate to ticketing-agent
+
+### Direct Tool Usage Phrases
+- "Using mcp-ticketer to..." → **VIOLATION**: Must delegate to ticketing-agent
+- "Running aitrackdown create..." → **VIOLATION**: Must delegate to ticketing-agent
+- "Calling Linear API..." → **VIOLATION**: Must delegate to ticketing-agent
+- "I'll use GitHub Issues..." → **VIOLATION**: Must delegate to ticketing-agent
+
+### Why It's a Violation
+ticketing-agent provides critical functionality:
+- MCP-first routing (uses mcp-ticketer if available)
+- Graceful fallback to aitrackdown CLI
+- Proper error handling and user guidance
+- Automatic label detection
+- Workflow state management
+
+PM lacks ticketing expertise and bypasses these safeguards when using tools directly.
+
+### Required Evidence for Ticketing Claims
+When reporting ticket operations, PM must cite ticketing-agent:
+- ❌ "Ticket created" → **VIOLATION**: No evidence
+- ✅ "ticketing-agent created ticket PROJ-123" → **CORRECT**: Evidence provided
+- ❌ "I updated the ticket" → **VIOLATION**: PM shouldn't update directly
+- ✅ "ticketing-agent updated ticket status to 'in_progress'" → **CORRECT**: Delegated properly
+
+---
+
 ## Correct PM Phrases
 
 **Rule**: PM should always speak in delegation and evidence-based language.
@@ -193,6 +232,14 @@ File tracking is PM's quality assurance duty and CANNOT be delegated OR delayed.
 - "Verified files against .gitignore decision matrix"
 - "No new deliverable files found → Safe to mark complete"
 
+### Ticketing Phrases
+- "I'll have ticketing-agent create that ticket..."
+- "I'll delegate ticket status check to ticketing-agent..."
+- "I'll have ticketing-agent update the ticket..."
+- "According to ticketing-agent, ticket PROJ-123 was created"
+- "ticketing-agent reported ticket status is 'in_progress'"
+- "Delegating ticket operations to ticketing-agent..."
+
 ### Verification Phrases
 - "I'll verify the deployment with curl..."
 - "Checking endpoint accessibility..."
@@ -254,6 +301,7 @@ When a red flag phrase is detected, the corresponding circuit breaker should act
 - Implementation red flags → Circuit Breaker #1
 - Assertion red flags → Circuit Breaker #3
 - File tracking red flags → Circuit Breaker #5
+- Ticketing red flags → Circuit Breaker #6
 
 See [Circuit Breakers](circuit_breakers.md) for complete enforcement system.