UNPKG

aiwg

Version:

Deployment tool and support utility for AI context. Copies agents, skills, commands, rules, and behaviors into the paths each AI platform reads (Claude Code, Codex, Copilot, Cursor, Warp, OpenClaw, and 6 more) so one source of truth works across 10 platfo

421 lines (319 loc) 16.5 kB
# Simple Language Translations **Purpose**: Map natural language user requests to SDLC flow orchestration templates. ## Core Principle **Users don't type slash commands. They use natural language.** As the core orchestrator, you interpret user intent and map it to appropriate flow templates, then coordinate multi-agent workflows. ## Translation Patterns ### Phase Transitions | User Says | Intent | Flow Template | Expected Duration | |-----------|--------|---------------|-------------------| | "Let's transition to Elaboration" | Start Inception→Elaboration | `flow-inception-to-elaboration` | 15-20 min | | "Move to Elaboration" | Same as above | `flow-inception-to-elaboration` | 15-20 min | | "Start Elaboration phase" | Same as above | `flow-inception-to-elaboration` | 15-20 min | | "Transition to Construction" | Start Elaboration→Construction | `flow-elaboration-to-construction` | 10-15 min | | "Begin Construction" | Same as above | `flow-elaboration-to-construction` | 10-15 min | | "Move to Transition phase" | Start Construction→Transition | `flow-construction-to-transition` | 10-15 min | | "Ready to deploy" | Same as above | `flow-construction-to-transition` | 10-15 min | | "Start Inception" | Concept→Inception | `flow-concept-to-inception` | 10-15 min | | "Begin the project" | Same as above | `flow-concept-to-inception` | 10-15 min | ### Workflow Requests | User Says | Intent | Flow Template | Expected Duration | |-----------|--------|---------------|-------------------| | "Run iteration {N}" | Execute dual-track iteration | `flow-iteration-dual-track` | 20-30 min | | "Start iteration {N}" | Same as above | `flow-iteration-dual-track` | 20-30 min | | "Discovery for iteration {N}" | Discovery track only | `flow-discovery-track` | 10-15 min | | "Delivery for iteration {N}" | Delivery track only | `flow-delivery-track` | 10-15 min | | "Deploy to production" | Production deployment | `flow-deploy-to-production` | 15-20 min | | "Start deployment" | Same as above | `flow-deploy-to-production` | 15-20 min | | "Run hypercare" | Post-launch monitoring | `flow-hypercare-monitoring` | 5-10 min setup | | "Start monitoring" | Same as above | `flow-hypercare-monitoring` | 5-10 min setup | ### Review Cycles | User Says | Intent | Flow Template | Expected Duration | |-----------|--------|---------------|-------------------| | "Start security review" | Security validation cycle | `flow-security-review-cycle` | 15-20 min | | "Run security check" | Same as above | `flow-security-review-cycle` | 15-20 min | | "Validate security" | Same as above | `flow-security-review-cycle` | 15-20 min | | "Run tests" | Execute test strategy | `flow-test-strategy-execution` | 10-15 min | | "Execute test suite" | Same as above | `flow-test-strategy-execution` | 10-15 min | | "Validate tests" | Same as above | `flow-test-strategy-execution` | 10-15 min | | "Check compliance" | Compliance validation | `flow-compliance-validation` | 15-20 min | | "Validate {framework} compliance" | Same as above | `flow-compliance-validation` | 15-20 min | | "Performance review" | Performance optimization | `flow-performance-optimization` | 10-15 min | | "Optimize performance" | Same as above | `flow-performance-optimization` | 10-15 min | ### Artifact Generation | User Says | Intent | Action | Expected Duration | |-----------|--------|--------|-------------------| | "Create architecture baseline" | Generate SAD + ADRs | Extract from `flow-inception-to-elaboration` Step 3 | 10-15 min | | "Generate SAD" | Software Architecture Document | Extract from `flow-inception-to-elaboration` Step 3 | 8-10 min | | "Create ADRs" | Architecture Decision Records | Extract from `flow-inception-to-elaboration` Step 3 | 5-8 min | | "Build test plan" | Master Test Plan | Extract from `flow-elaboration-to-construction` | 8-10 min | | "Create deployment plan" | Deployment planning | Extract from `flow-construction-to-transition` | 8-10 min | | "Generate risk register" | Initial risk list | Already done in intake, or `flow-risk-management-cycle` | 5-8 min | ### Status and Gates | User Says | Intent | Flow Template | Expected Duration | |-----------|--------|---------------|-------------------| | "Where are we?" | Project status | `project-status` | Instant | | "What's next?" | Same as above | `project-status` | Instant | | "Check project status" | Same as above | `project-status` | Instant | | "Can we transition?" | Gate check | `flow-gate-check {phase}` | 2-3 min | | "Are we ready for {phase}?" | Same as above | `flow-gate-check {phase}` | 2-3 min | | "Validate gate criteria" | Same as above | `flow-gate-check {phase}` | 2-3 min | | "Check handoff" | Handoff validation | `flow-handoff-checklist` | 3-5 min | | "Validate handoff" | Same as above | `flow-handoff-checklist` | 3-5 min | ### Risk and Change Management | User Says | Intent | Flow Template | Expected Duration | |-----------|--------|---------------|-------------------| | "Update risks" | Risk management cycle | `flow-risk-management-cycle` | 8-10 min | | "Review risks" | Same as above | `flow-risk-management-cycle` | 8-10 min | | "Manage risks" | Same as above | `flow-risk-management-cycle` | 8-10 min | | "Submit change request" | Change control | `flow-change-control` | 5-8 min | | "Process change" | Same as above | `flow-change-control` | 5-8 min | | "Evolve architecture" | Architecture evolution | `flow-architecture-evolution` | 10-15 min | | "Update architecture" | Same as above | `flow-architecture-evolution` | 10-15 min | | "Refine requirements" | Requirements evolution | `flow-requirements-evolution` | 8-10 min | | "Update requirements" | Same as above | `flow-requirements-evolution` | 8-10 min | ### Team Coordination | User Says | Intent | Flow Template | Expected Duration | |-----------|--------|---------------|-------------------| | "Onboard {name}" | Team member onboarding | `flow-team-onboarding` | 10-15 min | | "Add team member" | Same as above | `flow-team-onboarding` | 10-15 min | | "Knowledge transfer" | Knowledge transfer workflow | `flow-knowledge-transfer` | 10-15 min | | "Handoff to {name}" | Same as above | `flow-knowledge-transfer` | 10-15 min | | "Sync with {team}" | Cross-team coordination | `flow-cross-team-sync` | 8-10 min | | "Coordinate teams" | Same as above | `flow-cross-team-sync` | 8-10 min | | "Run retrospective" | Retrospective cycle | `flow-retrospective-cycle` | 10-15 min | | "Hold retro" | Same as above | `flow-retrospective-cycle` | 10-15 min | ### Incident and Operations | User Says | Intent | Flow Template | Expected Duration | |-----------|--------|---------------|-------------------| | "Handle incident" | Incident response | `flow-incident-response` | 15-20 min | | "Respond to incident" | Same as above | `flow-incident-response` | 15-20 min | | "Production issue" | Same as above | `flow-incident-response` | 15-20 min | ## Interpretation Guidelines ### 1. Context Awareness **Consider current project phase** when interpreting ambiguous requests: ```python # Example logic for "start" if user_says("start") and current_phase == "Inception Ready": → flow-concept-to-inception elif user_says("start") and current_phase == "Inception Active": → Clarify: "Start what? (Elaboration transition, security review, risk management)" elif user_says("start") and has_recent_incident: → flow-incident-response ``` ### 2. Clarification When Needed **Ask for clarification** if intent is ambiguous: ``` User: "Run security" You: "I can help with security. Did you mean: 1. Start security review cycle (flow-security-review-cycle) 2. Run security gate check (flow-gate-check) 3. Something else?" ``` ### 3. Extract Parameters **Parse relevant parameters** from natural language: ``` User: "Run iteration 5 with focus on performance" You interpret: - Flow: flow-iteration-dual-track - Parameter: iteration=5 - Guidance: "focus on performance" ``` ### 4. Offer Guidance Mode **Suggest --guidance or --interactive** for complex requests: ``` User: "Start security review" You: "I'll orchestrate a security review cycle. This will take ~15-20 minutes. Would you like to: 1. Proceed with standard security review 2. Provide guidance upfront (--guidance: priorities, constraints, focus) 3. Interactive mode (I'll ask 5-8 strategic questions) [If no response in 3 seconds, proceed with #1]" ``` ## Response Patterns ### Confirmation Pattern **Always confirm understanding** before starting orchestration: ``` User: "Let's transition to Elaboration" You: "Understood. I'll orchestrate the Inception → Elaboration transition. This will generate: - Software Architecture Document (SAD) - Architecture Decision Records (3-5 ADRs) - Master Test Plan - Elaboration Phase Plan I'll coordinate multiple agents for comprehensive review. Expected duration: 15-20 minutes. Starting orchestration... ``` ### Progress Update Pattern **Update user throughout** with clear indicators: ``` ✓ = Complete ⏳ = In progress ❌ = Error/blocked ⚠️ = Warning/attention needed Example: ✓ Initialized workspaces ⏳ SAD Draft (Architecture Designer)... ✓ SAD v0.1 draft complete (3,245 words) ⏳ Launching parallel review (4 agents)... ``` ### Completion Pattern **Summarize results** with actionable next steps: ``` ───────────────────────────────────────────── Transition Complete ✓ ───────────────────────────────────────────── Generated Artifacts: ✓ Software Architecture Document (.aiwg/architecture/software-architecture-doc.md) ✓ ADR-001: Database Selection (.aiwg/architecture/adr/ADR-001-database-selection.md) ✓ ADR-002: API Architecture (.aiwg/architecture/adr/ADR-002-api-architecture.md) ✓ Master Test Plan (.aiwg/testing/master-test-plan.md) ✓ Elaboration Phase Plan (.aiwg/planning/phase-plan-elaboration.md) Review Status: All artifacts reviewed and APPROVED by responsible roles Next Steps: - Review generated artifacts - Run "check if we can transition" to validate Elaboration readiness - Begin Elaboration activities per phase plan ``` ## Special Cases ### Intake Requests | User Says | Intent | Action | |-----------|--------|--------| | "Start new project" | Begin intake | `intake-wizard` with interactive mode | | "Analyze codebase" | Generate intake from code | `intake-from-codebase` | | "Complete intake" | Validate and start Inception | `intake-start` | ### Health and Status | User Says | Intent | Action | |-----------|--------|--------| | "Project health" | Overall health check | `project-health-check` | | "Build report" | Artifact index | `build-artifact-index` | | "Check traceability" | Verify traceability | `check-traceability` | ### PoC and Spikes | User Says | Intent | Action | |-----------|--------|--------| | "Build PoC for {concept}" | Create proof of concept | `build-poc` | | "Validate {technical-risk}" | Same as above | `build-poc` | | "Spike {feature}" | Same as above | `build-poc` | ### Help and Troubleshooting | User Says | Intent | Action | |-----------|--------|--------| | "Help with AIWG" | General help | `aiwg-kb "help"` | | "How do I fix {issue}?" | Troubleshooting | `aiwg-kb "{issue}"` | | "AIWG not working" | Setup troubleshooting | `aiwg-kb "setup issues"` | | "Agent not found" | Deployment troubleshooting | `aiwg-kb "deployment issues"` | | "Template not found" | Path troubleshooting | `aiwg-kb "path issues"` | | "Why can't Claude find {x}?" | Troubleshooting | `aiwg-kb "deployment issues"` | | "{Platform} not working" | Platform troubleshooting | `aiwg-kb "platform issues"` | | "What commands can I use?" | Natural language reference | `aiwg-kb "natural language"` | | "How do I start?" | Quickstart | `aiwg-kb "quickstart"` | | "AIWG documentation" | Documentation | `aiwg-kb "help"` | ## Regex Patterns (For Implementation) If implementing pattern matching, use these regex patterns: ```python PHASE_TRANSITIONS = { r"(transition|move|start|begin).*(elaboration)": "flow-inception-to-elaboration", r"(transition|move|start|begin).*(construction)": "flow-elaboration-to-construction", r"(transition|move|start|begin).*(transition)": "flow-construction-to-transition", } WORKFLOWS = { r"(run|start|execute).*(iteration|sprint)\s*(\d+)": "flow-iteration-dual-track", r"(deploy|deployment|release).*(production|prod)": "flow-deploy-to-production", r"(security|sec).*(review|check|validate)": "flow-security-review-cycle", } STATUS = { r"(where|what).*(are we|next|status)": "project-status", r"(can|ready).*(transition|move|proceed)": "flow-gate-check", } HELP = { r"(help|fix|problem|issue|error).*(aiwg|install|setup)": "aiwg-kb setup", r"(agent|command).*(not found|missing|can't find)": "aiwg-kb deployment", r"(template|path|directory).*(not found|missing|error)": "aiwg-kb path", r"(claude|factory|warp).*(not working|issue|problem)": "aiwg-kb platform", r"(how|what).*(start|begin|commands?)": "aiwg-kb quickstart", r"(help|documentation|docs).*aiwg": "aiwg-kb help", } ``` ## Examples ### Example 1: Clear Intent ``` User: "Let's transition to Elaboration" Intent: Phase transition (clear) Flow: flow-inception-to-elaboration Parameters: None Guidance: None Action: Confirm and orchestrate ``` ### Example 2: Needs Clarification ``` User: "Start" Intent: Ambiguous Current Phase: Inception Active Response: "What would you like to start? - Elaboration transition (if Inception complete) - Security review cycle - Risk management cycle - Specific artifact generation" ``` ### Example 3: Extract Guidance ``` User: "Run security review with focus on authentication and HIPAA compliance" Intent: Security review cycle Flow: flow-security-review-cycle Parameters: None Guidance: "Focus on authentication and HIPAA compliance" Action: Confirm and orchestrate with guidance ``` ### Example 4: Interactive Suggestion ``` User: "Create architecture baseline" Intent: Generate SAD + ADRs Flow: Extract from flow-inception-to-elaboration Step 3 Response: "I'll generate the architecture baseline (SAD + ADRs). This is a comprehensive artifact. Would you like: 1. Standard generation (15 min) 2. Provide guidance first (priorities, constraints) 3. Interactive mode (I'll ask 7 questions) [Default to #1 if no response]" ``` ## Natural Language Command Reference **Full list of recognized phrases by category:** ### Phase Work - "transition to {phase}" | "move to {phase}" | "start {phase}" - "run iteration {N}" | "start iteration {N}" - "deploy to production" | "start deployment" ### Reviews and Validation - "security review" | "run security" | "validate security" - "run tests" | "execute tests" | "test suite" - "check compliance" | "validate compliance" - "performance review" | "optimize performance" ### Artifacts - "create {artifact}" | "generate {artifact}" | "build {artifact}" - "architecture baseline" | "SAD" | "ADRs" - "test plan" | "deployment plan" | "risk register" ### Status - "where are we" | "what's next" | "project status" - "can we transition" | "ready for {phase}" | "check gate" ### Team and Process - "onboard {name}" | "add team member" - "knowledge transfer" | "handoff to {name}" - "retrospective" | "retro" | "hold retro" ### Operations - "incident" | "production issue" | "handle incident" - "hypercare" | "monitoring" | "post-launch" ## Implementation Notes 1. **Pattern Matching**: Use fuzzy matching (not exact), prioritize recent context 2. **Default to Safety**: If unsure, ask for clarification rather than guessing 3. **Learn Over Time**: Track which phrases users use most, adapt patterns 4. **Context Stack**: Maintain conversation context to resolve "it", "that", "same" 5. **Abbreviations**: Support common abbreviations (e.g., "sec review", "perf test") ## Testing **Test scenarios for validation:** ``` ✓ "transition to Elaboration" → flow-inception-to-elaboration ✓ "run security review" → flow-security-review-cycle ✓ "where are we" → project-status ✓ "can we move forward" + (context: Inception) → flow-gate-check inception ✓ "start" + (phase: Pre-Inception) → intake-wizard ✓ "run iteration 5 with performance focus" → flow-iteration-dual-track (iteration=5, guidance="performance focus") ```