UNPKG

claude-flow-novice

Version:

Claude Flow Novice - Advanced orchestration platform for multi-agent AI workflows with CFN Loop architecture Includes CodeSearch (hybrid SQLite + pgvector), mem0/memgraph specialists, and all CFN skills.

358 lines (287 loc) 9.29 kB
--- description: "Parse natural language epic documents to structured JSON configuration" argument-hint: "<epic-directory> [--output <file>] [--validate]" allowed-tools: ["Read", "Write", "Bash", "Glob", "Grep"] --- # Parse Epic - Natural Language to Structured JSON Convert natural language epic markdown documents into structured JSON configuration for CFN Loop execution. **Input**: Natural language epic directory with markdown files **Output**: Structured JSON configuration with phases, sprints, dependencies ## Usage ```bash # Basic parsing /parse-epic planning/example-epic # With custom output file /parse-epic planning/my-epic --output epic-config.json # With validation /parse-epic planning/auth-epic --validate # Full options /parse-epic planning/example-epic --output custom-epic.json --validate ``` ## Command Arguments - `<epic-directory>`: **Required** - Path to epic directory containing markdown files - `--output <file>`: Optional - Custom output JSON file (default: `<epic-name>-config.json`) - `--validate`: Optional - Validate parsed configuration against schema ## Epic Directory Structure ``` planning/example-epic/ ├── OVERVIEW.md # Epic description, goals, scope ├── phase-1-authentication.md # Phase 1 details ├── phase-2-authorization.md # Phase 2 details ├── phase-3-session-mgmt.md # Phase 3 details └── dependencies.md # Cross-phase dependencies (optional) ``` ## Markdown Format ### OVERVIEW.md ```markdown # Auth System Epic **Goal**: Build complete authentication and authorization system ## Scope ### In Scope - JWT-based authentication - Role-based access control (RBAC) - Session management - Password security (bcrypt) ### Out of Scope - OAuth/social login - Multi-factor authentication - Biometric authentication ## Risk Profile public-facing-medium-risk ## Estimated Timeline 3 phases, 8 sprints total ``` ### Phase Markdown (phase-1-authentication.md) ```markdown # Phase 1: User Authentication **Phase ID**: 1 **Dependencies**: None ## Deliverables - Login API endpoint (POST /auth/login) - JWT token generation - Password validation - bcrypt password hashing ## Sprints ### Sprint 1.1: Core Login API - POST /auth/login endpoint - JWT token generation - Basic password validation **Estimated Agents**: 3 ### Sprint 1.2: Password Security - bcrypt password hashing - Salt generation - Hash verification **Estimated Agents**: 2 ## Success Criteria - All login tests passing - Security audit complete - API documentation updated ``` ## Output JSON Structure The parser generates a structured JSON configuration: ```json { "epic_name": "auth-system", "epic_goal": "Build complete authentication and authorization system", "scope": { "in_scope": [ "JWT-based authentication", "Role-based access control (RBAC)", "Session management", "Password security (bcrypt)" ], "out_of_scope": [ "OAuth/social login", "Multi-factor authentication", "Biometric authentication" ], "risk_profile": "public-facing-medium-risk" }, "phases": [ { "phase_id": "1", "phase_name": "User Authentication", "dependencies": [], "deliverables": [ "Login API endpoint (POST /auth/login)", "JWT token generation", "Password validation", "bcrypt password hashing" ], "sprints": [ { "sprint_id": "1.1", "sprint_name": "Core Login API", "deliverables": [ "POST /auth/login endpoint", "JWT token generation", "Basic password validation" ], "estimated_agents": 3 }, { "sprint_id": "1.2", "sprint_name": "Password Security", "deliverables": [ "bcrypt password hashing", "Salt generation", "Hash verification" ], "estimated_agents": 2 } ], "success_criteria": [ "All login tests passing", "Security audit complete", "API documentation updated" ] } ], "decision_authority_config": { "auto_approve_threshold": 0.90, "auto_relaunch_max_iteration": 10, "escalation_criteria": [ "security vulnerability", "data loss risk" ] } } ``` ## Validation When using `--validate`, the parser checks: - Valid phase IDs and naming - Dependency resolution (no circular dependencies) - Sprint numbering consistency (1.1, 1.2, etc.) - Deliverables are non-empty - Scope boundaries are defined - Risk profile is valid (internal-only-low-risk, public-facing-medium-risk, critical-high-risk) - Success criteria are specified **Validation Errors:** ``` Validation Failed: - Phase 2 dependency on Phase 5 (does not exist) - Sprint 1.3 missing deliverables - Risk profile "high" invalid (use: internal-only-low-risk, public-facing-medium-risk, critical-high-risk) ``` ## Integration with CFN Loop After parsing, use the JSON configuration with CFN Loop commands: ```bash # 1. Parse epic to JSON /parse-epic planning/auth-epic --output auth-config.json --validate # 2. Execute epic using parsed configuration /cfn-loop-epic "$(cat auth-config.json)" # Or manually reference phases /cfn-loop-sprints "Phase 1: User Authentication (from auth-config.json)" ``` ## Automatic Scope Storage The parser automatically stores scope boundaries in memory for Product Owner access: ```javascript // Automatically executed during parsing mcp__claude-flow-novice__memory_usage({ action: "store", namespace: "scope-control", key: "project-boundaries", value: JSON.stringify({ primary_goal: epicConfig.epic_goal, in_scope: epicConfig.scope.in_scope, out_of_scope: epicConfig.scope.out_of_scope, risk_profile: epicConfig.scope.risk_profile, decision_authority_config: epicConfig.decision_authority_config }) }) ``` ## Example Workflows ### Workflow 1: Parse and Execute Epic ```bash # Step 1: Create epic markdown files in planning/auth-epic/ # Step 2: Parse to JSON /parse-epic planning/auth-epic --output auth-config.json --validate # Step 3: Execute epic (automatic phase orchestration) /cfn-loop-epic "$(cat auth-config.json)" ``` ### Workflow 2: Parse and Execute Single Phase ```bash # Step 1: Parse epic /parse-epic planning/my-epic --validate # Step 2: Extract specific phase (manual) # Read my-epic-config.json, copy Phase 1 details # Step 3: Execute single phase /cfn-loop-sprints "Phase 1 content from parsed JSON" ``` ### Workflow 3: Validate Epic Structure ```bash # Validate epic structure before execution /parse-epic planning/complex-epic --validate # Review validation output # Fix any errors in markdown files # Re-run validation until clean ``` ## Output Files **Generated Files:** - `<epic-name>-config.json` - Structured epic configuration - `<epic-name>-validation-report.json` - Validation results (if --validate) **Example Output:** ```bash 🔍 Epic Parser - Natural Language to Structured JSON 📂 Epic directory: planning/auth-epic 📄 Overview file: OVERVIEW.md Parsed successfully: - Epic: auth-system - Phases: 3 - Total Sprints: 8 - Scope boundaries: - Dependencies: 📄 Output: auth-config.json Validation: PASSED Ready for CFN Loop execution! ``` ## Error Handling **Common Errors:** 1. **Missing OVERVIEW.md** ``` Epic directory must contain OVERVIEW.md ``` 2. **Invalid Phase Dependencies** ``` Phase 3 depends on Phase 5 (not found) ``` 3. **Circular Dependencies** ``` Circular dependency detected: Phase 1 Phase 2 Phase 1 ``` 4. **Invalid Sprint Numbering** ``` Sprint 1.5 follows 1.2 (expected 1.3) ``` ## Advanced Features ### Custom Risk Profiles ```markdown ## Risk Profile custom-high-security-critical ## Decision Authority - Auto-approve threshold: 0.95 - Max iteration: 5 - Escalation: security-team-approval-required ``` ### Cross-Phase Dependencies Create `dependencies.md`: ```markdown # Epic Dependencies - Phase 2 requires Phase 1 (authentication must complete first) - Phase 3 requires Phase 1, Phase 2 (session needs auth + authorization) - Phase 4 requires Phase 3 (admin panel needs sessions) ``` ### Sprint Estimation ```markdown ### Sprint 2.3: Complex Authorization Logic **Estimated Agents**: 8 **Complexity**: High **Rationale**: Multi-tenant RBAC with custom permission system ``` ## Tips for Epic Creation 1. **Start with OVERVIEW.md** - Define clear scope boundaries 2. **One phase per file** - Keep phase definitions modular 3. **Explicit dependencies** - Document all cross-phase requirements 4. **Realistic sprint sizing** - 2-5 deliverables per sprint 5. **Validate early** - Use `--validate` during epic creation 6. **Clear success criteria** - Define measurable phase completion criteria