mirror of
https://github.com/catlog22/Claude-Code-Workflow.git
synced 2026-03-11 17:21:03 +08:00
- Add CCW-COMMANDS-SKILLS-REFERENCE.md with comprehensive documentation - Update docs/reference/commands-skills.md (English) - Update docs/zh/reference/commands-skills.md (Chinese) - Include 55 Slash Commands and 41 Skills - Each entry follows 8-field format: Who, When, What, How, Output, Done, Failure & Recovery, Impact & Boundary - Exclude academic skills (scholar-*, journal-paper-writing, software-manual)
51 KiB
51 KiB
CCW Commands & Skills Reference Manual
Generated: 2026-03-08 | Project: Claude Code Workspace (CCW)
This reference manual provides quick lookup for all slash commands and skills in the CCW system.
Table of Contents
- Section 1: Slash Commands
- Section 2: Skills
- Appendix A: Quick Reference Table
- Appendix B: Command Chain Examples
Section 1: Slash Commands
Core
/ccw
| Field | Description |
|---|---|
| Who | Developer starting any task |
| When | Entry point for workflow orchestration |
| What | Main workflow orchestrator that analyzes intent, dispatches to appropriate skill chain. Example: /ccw "implement user authentication" |
| How | Parses intent -> Selects skill chain -> Executes orchestrator |
| Output | Delegates to workflow-plan, issue-manage, or other skills |
| Done | Appropriate skill completes execution |
| Failure & Recovery | Fallback to manual skill selection if intent unclear |
| Impact & Boundary | Project-wide orchestration entry point |
/ccw-coordinator
| Field | Description |
|---|---|
| Who | System needing command orchestration |
| When | Multi-command coordination required |
| What | Command orchestration tool that analyzes requirements and dispatches commands. Example: /ccw-coordinator "review and fix auth module" |
| How | Requirement analysis -> Command selection -> Sequential execution |
| Output | Coordinated command execution results |
| Done | All dispatched commands complete |
| Failure & Recovery | Partial results preserved, user notified of failures |
| Impact & Boundary | Cross-command coordination |
/flow-create
| Field | Description |
|---|---|
| Who | Developer creating reusable workflow templates |
| When | Need to document and share workflow patterns |
| What | Flow template generator for creating reusable command chains. Example: /flow-create "auth-feature-flow" |
| How | Template definition -> Parameter extraction -> Flow file generation |
| Output | Flow template file in .claude/flows/ |
| Done | Template file created and validated |
| Failure & Recovery | Validation errors show specific issues |
| Impact & Boundary | Workflow template creation only |
DDD (Document-Driven Development)
/ddd:scan
| Field | Description |
|---|---|
| Who | Developer onboarding to existing project without specs |
| When | Starting doc-driven workflow on legacy codebase |
| What | Scans codebase to build doc-index by reverse-engineering structure. Example: /ddd:scan -y "existing project" |
| How | Framework detection -> Component discovery -> Feature inference -> Requirement extraction -> doc-index.json |
| Output | .workflow/.doc-index/doc-index.json + feature-maps/ + tech-registry/ |
| Done | doc-index.json created with components, features, IREQ entries |
| Failure & Recovery | Graceful degradation if DeepWiki unavailable |
| Impact & Boundary | Code-first entry point, can upgrade to spec-first later |
/ddd:index-build
| Field | Description |
|---|---|
| Who | Developer with existing spec-generator outputs |
| When | Building index from REQ, ADR, EPIC documents |
| What | Builds doc-index from spec outputs and maps entities to code. Example: /ddd:index-build --spec SPEC-auth-2026-03-08 |
| How | Parse specs -> Codebase mapping via CLI -> Feature grouping -> doc-index.json |
| Output | .workflow/.doc-index/doc-index.json with spec-first traceability |
| Done | All spec entities linked to code components |
| Failure & Recovery | Orphan components flagged for review |
| Impact & Boundary | Spec-first entry point, merges with existing code-first index |
/ddd:plan
| Field | Description |
|---|---|
| Who | Developer planning document-driven task |
| When | Before any development task in DDD workflow |
| What | Queries doc-index, explores codebase with doc-aware angles, produces plan.json. Example: /ddd:plan -y "add rate limiting" |
| How | Doc-index query -> Guided exploration -> Clarification -> Task planning -> Handoff |
| Output | .workflow/.doc-index/planning/{session}/plan.json + TASK-*.json |
| Done | Plan with doc_context traceability generated |
| Failure & Recovery | Schema version mismatch warns but continues |
| Impact & Boundary | Requires existing doc-index.json |
/ddd:execute
| Field | Description |
|---|---|
| Who | Developer executing document-aware tasks |
| When | After /ddd:plan produces plan.json |
| What | Document-aware execution engine with doc-context loading. Example: /ddd:execute --in-memory |
| How | Load plan -> Read doc-context -> Execute task -> Update doc-index |
| Output | Code changes + updated doc-index traceability |
| Done | All tasks executed, doc-index updated |
| Failure & Recovery | Task failures logged, recovery options presented |
| Impact & Boundary | Requires plan.json from /ddd:plan |
/ddd:doc-generate
| Field | Description |
|---|---|
| Who | Developer regenerating documentation |
| When | After doc-index built or needs refresh |
| What | Generates full document tree from doc-index.json. Example: /ddd:doc-generate --force --layer all |
| How | Layer 3 (components) -> Layer 2 (features) -> Layer 1 (indexes) |
| Output | tech-registry/.md, feature-maps/.md, README.md, ARCHITECTURE.md |
| Done | All layers generated, SCHEMA.md created |
| Failure & Recovery | Existing docs warning, --force to overwrite |
| Impact & Boundary | Read-only from doc-index, generates MD files |
/ddd:doc-refresh
| Field | Description |
|---|---|
| Who | Developer after incremental code changes |
| When | Code changed, need targeted doc updates |
| What | Incrementally updates affected documents after code changes. Example: /ddd:doc-refresh -y |
| How | Detect changed files -> Find affected docs -> Regenerate only affected |
| Output | Updated subset of documentation files |
| Done | Affected docs updated, unchanged docs preserved |
| Failure & Recovery | Falls back to full regeneration if detection fails |
| Impact & Boundary | Incremental update, more efficient than full regenerate |
/ddd:sync
| Field | Description |
|---|---|
| Who | Developer after task completion |
| When | Post-task synchronization to update index |
| What | Post-task synchronization updates doc-index with completed work. Example: /ddd:sync -y "auth feature complete" |
| How | Detect changes -> Update components -> Update requirements status -> Add action log |
| Output | Updated doc-index.json with action entries |
| Done | Index reflects current codebase state |
| Failure & Recovery | Merge conflicts resolved automatically |
| Impact & Boundary | Post-execution sync only |
/ddd:update
| Field | Description |
|---|---|
| Who | Developer maintaining doc-index |
| When | Detecting drift between code and docs |
| What | Incremental index update detecting code changes. Example: /ddd:update --scope src/auth |
| How | File hash comparison -> Detect changes -> Update affected entries |
| Output | Updated doc-index.json entries |
| Done | Index entries match current code |
| Failure & Recovery | Unchanged files skipped |
| Impact & Boundary | Lightweight incremental update |
/ddd:auto
| Field | Description |
|---|---|
| Who | Developer wanting full automation |
| When | Quick feature implementation with docs |
| What | Chain command for automated DDD workflow. Example: /ddd:auto -y "implement search feature" |
| How | plan -> execute -> sync in single command |
| Output | Complete implementation + updated docs |
| Done | Feature implemented and documented |
| Failure & Recovery | Stops at first failing phase |
| Impact & Boundary | Full automation, less control |
Issue
/issue:new
| Field | Description |
|---|---|
| Who | Developer creating structured issue |
| When | Starting issue tracking for a task |
| What | Creates structured issue from GitHub URL or description. Example: /issue:new "Fix login timeout bug" |
| How | Parse input -> Structure issue -> Create in issues.jsonl |
| Output | Issue entry in .workflow/issues/issues.jsonl |
| Done | Issue created with status "registered" |
| Failure & Recovery | Duplicate detection warns user |
| Impact & Boundary | Issue creation only, no planning |
/issue:discover
| Field | Description |
|---|---|
| Who | Developer finding potential issues |
| When | Starting issue management workflow |
| What | Discovers potential issues from multiple sources. Example: /issue:discover --scope src/auth |
| How | Scan codebase -> Detect patterns -> Generate issue candidates |
| Output | List of discovered issue candidates |
| Done | Candidates presented for user selection |
| Failure & Recovery | Empty results suggest manual creation |
| Impact & Boundary | Discovery only, no automatic creation |
/issue:discover-by-prompt
| Field | Description |
|---|---|
| Who | Developer with natural language query |
| When | Finding issues via description |
| What | Discovers issues from user prompt with Gemini analysis. Example: /issue:discover-by-prompt "authentication problems" |
| How | Gemini analysis -> Pattern matching -> Issue suggestions |
| Output | Issue candidates matching prompt |
| Done | Candidates presented for confirmation |
| Failure & Recovery | Falls back to keyword search |
| Impact & Boundary | Natural language discovery |
/issue:from-brainstorm
| Field | Description |
|---|---|
| Who | Developer converting ideas to issues |
| When | After brainstorm session |
| What | Converts brainstorm session ideas into issues. Example: /issue:from-brainstorm BRAIN-2026-03-08 |
| How | Parse brainstorm -> Extract actionables -> Create issues |
| Output | Multiple issues from brainstorm artifacts |
| Done | All actionables converted to issues |
| Failure & Recovery | Partial conversion on parse errors |
| Impact & Boundary | Requires completed brainstorm session |
/issue:plan
| Field | Description |
|---|---|
| Who | Developer planning issue resolution |
| When | Issue registered, needs solution |
| What | Batch plans issue resolution using issue-planning-agent. Example: /issue:plan ISS-001 ISS-002 |
| How | Load issues -> Generate solutions -> Bind to issues |
| Output | Solution files in .workflow/issues/solutions/ |
| Done | Issues have bound solutions |
| Failure & Recovery | Partial planning on complex issues |
| Impact & Boundary | Planning only, no execution |
/issue:queue
| Field | Description |
|---|---|
| Who | Developer forming execution queue |
| When | Solutions planned, ready to execute |
| What | Forms execution queue from bound solutions. Example: /issue:queue ISS-001 ISS-002 |
| How | Build DAG -> Resolve dependencies -> Create queue |
| Output | Queue file in .workflow/issues/queues/ |
| Done | Queue ready for execution |
| Failure & Recovery | Circular dependency detection |
| Impact & Boundary | Queue formation only |
/issue:execute
| Field | Description |
|---|---|
| Who | Developer executing queued issues |
| When | Queue formed, ready to run |
| What | Executes queue with DAG-based parallel orchestration. Example: /issue:execute -y |
| How | DAG traversal -> Parallel execution -> Progress tracking |
| Output | Task execution results, code changes |
| Done | All queued tasks completed |
| Failure & Recovery | Failed tasks skipped, dependents notified |
| Impact & Boundary | Execution engine for issue workflow |
/issue:convert-to-plan
| Field | Description |
|---|---|
| Who | Developer with existing planning artifacts |
| When | Converting lite-plan or brainstorm to issue |
| What | Converts planning artifacts to issue format. Example: /issue:convert-to-plan plan.json |
| How | Parse artifact -> Extract tasks -> Create issue + solution |
| Output | Issue and solution from plan |
| Done | Plan converted to trackable issue |
| Failure & Recovery | Partial conversion preserved |
| Impact & Boundary | Artifact conversion only |
IDAW (Iterative Development Automation Workflow)
/idaw:status
| Field | Description |
|---|---|
| Who | Developer checking IDAW progress |
| When | During IDAW session |
| What | Views IDAW task and session progress. Example: /idaw:status |
| How | Read session state -> Display progress |
| Output | Status summary of active IDAW session |
| Done | Status displayed |
| Failure & Recovery | No active session message |
| Impact & Boundary | Read-only status check |
/idaw:add
| Field | Description |
|---|---|
| Who | Developer adding IDAW tasks |
| When | Building IDAW task list |
| What | Adds IDAW tasks via manual creation or import. Example: /idaw:add "Implement caching layer" |
| How | Parse input -> Create task entry -> Add to session |
| Output | Task added to IDAW session |
| Done | Task registered in session |
| Failure & Recovery | Duplicate task warning |
| Impact & Boundary | Task creation only |
/idaw:run
| Field | Description |
|---|---|
| Who | Developer starting IDAW execution |
| When | Tasks added, ready to run |
| What | IDAW orchestrator executing task skill chain. Example: /idaw:run -y |
| How | Task dispatch -> Skill execution -> Progress tracking |
| Output | Task execution results |
| Done | All tasks processed |
| Failure & Recovery | Failed tasks logged, continue others |
| Impact & Boundary | Main IDAW execution engine |
/idaw:run-coordinate
| Field | Description |
|---|---|
| Who | Developer needing coordinated execution |
| When | Complex multi-task IDAW session |
| What | IDAW coordinator executing task skill chain. Example: /idaw:run-coordinate |
| How | Coordinate multiple task executions -> Handle dependencies |
| Output | Coordinated execution results |
| Done | All coordinated tasks complete |
| Failure & Recovery | Dependency failures handled |
| Impact & Boundary | Coordinated execution mode |
/idaw:resume
| Field | Description |
|---|---|
| Who | Developer resuming interrupted IDAW |
| When | IDAW session interrupted |
| What | Resumes interrupted IDAW session from last checkpoint. Example: /idaw:resume |
| How | Load state -> Resume from checkpoint -> Continue execution |
| Output | Continued execution from interruption point |
| Done | Session completes |
| Failure & Recovery | Checkpoint validation on load |
| Impact & Boundary | Resume capability for IDAW |
Workflow
/workflow:session:start
| Field | Description |
|---|---|
| Who | Developer starting workflow session |
| When | Beginning any workflow work |
| What | Discovers or starts workflow session with conflict detection. Example: /workflow:session:start --auto "implement OAuth2" |
| How | Check active -> Detect conflicts -> Create/reuse session |
| Output | SESSION_ID: WFS-{slug} |
| Done | Session ready for work |
| Failure & Recovery | Multiple sessions warning |
| Impact & Boundary | Session initialization, calls /workflow:spec:setup if needed |
/workflow:session:list
| Field | Description |
|---|---|
| Who | Developer reviewing sessions |
| When | Checking available sessions |
| What | Lists all workflow sessions with status filtering. Example: /workflow:session:list --status active |
| How | Scan .workflow/active -> Parse session metadata -> Display |
| Output | Session list with status, dates |
| Done | List displayed |
| Failure & Recovery | Empty list if no sessions |
| Impact & Boundary | Read-only listing |
/workflow:session:sync
| Field | Description |
|---|---|
| Who | Developer syncing session work |
| When | After completing work in session |
| What | Quick-sync session work to specs/*.md and project-tech. Example: /workflow:session:sync -y "auth complete" |
| How | Gather context -> Extract updates -> Write specs + project-tech |
| Output | Updated specs/*.md and project-tech.json |
| Done | Sync complete message |
| Failure & Recovery | Duplicate rules skipped silently |
| Impact & Boundary | Updates project guidelines |
/workflow:session:resume
| Field | Description |
|---|---|
| Who | Developer resuming paused session |
| When | Returning to paused work |
| What | Resumes most recently paused workflow session. Example: /workflow:session:resume |
| How | Find paused session -> Load state -> Continue |
| Output | Session context restored |
| Done | Ready to continue work |
| Failure & Recovery | No paused session error |
| Impact & Boundary | Session resumption |
/workflow:session:complete
| Field | Description |
|---|---|
| Who | Developer finishing session |
| When | Session work complete |
| What | Archives completed session and removes from active. Example: /workflow:session:complete -y |
| How | Finalize state -> Archive -> Remove from active |
| Output | Archived session in .workflow/archived/ |
| Done | Session archived, specs updated |
| Failure & Recovery | Pending tasks warning |
| Impact & Boundary | Session finalization |
/workflow:spec:setup
| Field | Description |
|---|---|
| Who | Developer initializing project specs |
| When | First time setup or regenerating |
| What | Initializes project-level state via cli-explore-agent analysis. Example: /workflow:spec:setup --regenerate |
| How | cli-explore-agent -> Generate project-tech.json -> Interactive questionnaire -> Write specs |
| Output | .workflow/project-tech.json + .ccw/specs/*.md |
| Done | Project specs configured |
| Failure & Recovery | Agent failure falls back to basic init |
| Impact & Boundary | Project-level initialization |
/workflow:spec:add
| Field | Description |
|---|---|
| Who | Developer adding project rules |
| When | Adding conventions, constraints, or learnings |
| What | Adds specs, conventions, constraints to project guidelines. Example: /workflow:spec:add -y "Use functional components" |
| How | Parse input -> Auto-detect type -> Append to spec file |
| Output | Updated spec file in .ccw/specs/ |
| Done | Rule added confirmation |
| Failure & Recovery | Duplicate detection |
| Impact & Boundary | Single rule addition |
/workflow:clean
| Field | Description |
|---|---|
| Who | Developer cleaning code |
| When | Before commits or after refactoring |
| What | Intelligent code cleanup with mainline detection. Example: /workflow:clean -y |
| How | Detect unused -> Remove dead code -> Format |
| Output | Cleaned code files |
| Done | Cleanup report displayed |
| Failure & Recovery | Preserves files with unclear usage |
| Impact & Boundary | Code cleanup only |
/workflow:debug-with-file
| Field | Description |
|---|---|
| Who | Developer debugging issues |
| When | Encountering bugs or errors |
| What | Interactive hypothesis-driven debugging with file output. Example: /workflow:debug-with-file "login fails on timeout" |
| How | Hypothesis generation -> Evidence collection -> Validation -> Fix |
| Output | Debug report + potential fixes |
| Done | Root cause identified or fix applied |
| Failure & Recovery | Multiple hypothesis iteration |
| Impact & Boundary | Debugging workflow |
/workflow:analyze-with-file
| Field | Description |
|---|---|
| Who | Developer analyzing code |
| When | Understanding codebase or issues |
| What | Interactive collaborative analysis with multi-CLI support. Example: /workflow:analyze-with-file "analyze auth flow" |
| How | Discovery -> Analysis -> Synthesis -> Report |
| Output | Analysis report in session folder |
| Done | Analysis complete with insights |
| Failure & Recovery | Partial analysis on errors |
| Impact & Boundary | Analysis workflow |
/workflow:brainstorm-with-file
| Field | Description |
|---|---|
| Who | Developer exploring solutions |
| When | Starting new feature or solving complex problem |
| What | Interactive brainstorming with multi-CLI support. Example: /workflow:brainstorm-with-file "caching strategy" |
| How | Multi-perspective analysis -> Idea synthesis -> Recommendations |
| Output | Brainstorm report with actionables |
| Done | Ideas documented and prioritized |
| Failure & Recovery | Partial results preserved |
| Impact & Boundary | Ideation workflow |
/workflow:collaborative-plan-with-file
| Field | Description |
|---|---|
| Who | Developer planning collaboratively |
| When | Complex feature planning |
| What | Collaborative planning with Plan Note output. Example: /workflow:collaborative-plan-with-file "user dashboard redesign" |
| How | Context gathering -> Planning -> Plan Note generation |
| Output | Plan with structured notes |
| Done | Plan ready for execution |
| Failure & Recovery | Iterative refinement |
| Impact & Boundary | Planning workflow |
/workflow:integration-test-cycle
| Field | Description |
|---|---|
| Who | Developer running integration tests |
| When | After implementation, before merge |
| What | Self-iterating integration test workflow with reflection log. Example: /workflow:integration-test-cycle -y "test auth flow" |
| How | Generate tests -> Execute -> Reflect -> Iterate until pass |
| Output | Passing tests + reflection-log.md |
| Done | All tests passing |
| Failure & Recovery | Reflection-driven adjustment loop |
| Impact & Boundary | Integration testing workflow |
/workflow:refactor-cycle
| Field | Description |
|---|---|
| Who | Developer addressing tech debt |
| When | Systematic code improvement |
| What | Tech debt discovery and self-iterating refactoring. Example: /workflow:refactor-cycle -y "src/auth module" |
| How | Debt discovery -> Prioritization -> Refactor -> Validate -> Reflect |
| Output | reflection-log.md + state.json + code changes |
| Done | All queued debt items processed |
| Failure & Recovery | Per-item validation with rollback |
| Impact & Boundary | Tech debt lifecycle management |
/workflow:roadmap-with-file
| Field | Description |
|---|---|
| Who | Developer planning strategic requirements |
| When | Large feature or project planning |
| What | Strategic requirement roadmap with iterative decomposition. Example: /workflow:roadmap-with-file -y "implement auth system" |
| How | Understand -> Decompose -> Iterate -> Validate -> Handoff |
| Output | roadmap.md + issues.jsonl |
| Done | Roadmap approved, issues ready |
| Failure & Recovery | Max iteration rounds force proceed |
| Impact & Boundary | Strategic planning workflow |
/workflow:unified-execute-with-file
| Field | Description |
|---|---|
| Who | Developer executing plans |
| When | After planning, ready to implement |
| What | Universal execution engine for any planning output. Example: /workflow:unified-execute-with-file -y plan.json |
| How | Load plan -> Validate -> Wave execution -> Track progress |
| Output | Code changes + execution-events.md |
| Done | All tasks executed |
| Failure & Recovery | Per-task retry with rollback |
| Impact & Boundary | Universal execution engine |
UI Design Commands
| Command | Who | When | What | Output |
|---|---|---|---|---|
/workflow:ui-design:explore-auto |
Designer/Developer | Starting UI design | Interactive exploratory workflow | Style variants + layout templates + prototypes |
/workflow:ui-design:imitate-auto |
Designer/Developer | Imitating reference | UI design imitation workflow | Prototypes matching reference style |
/workflow:ui-design:generate |
Designer/Developer | Assembling prototypes | Pure assembler combining layouts + tokens | HTML/CSS prototypes |
/workflow:ui-design:style-extract |
Designer/Developer | Extracting styles | Extract design tokens from images | design-tokens.json + style-guide.md |
/workflow:ui-design:layout-extract |
Designer/Developer | Extracting layouts | Extract layout templates | layout-*.json files |
/workflow:ui-design:animation-extract |
Designer/Developer | Extracting animations | Extract animation patterns | animation-tokens.json |
/workflow:ui-design:codify-style |
Developer | From code to design | Extract styles from code files | Design tokens from CSS/SCSS |
/workflow:ui-design:import-from-code |
Developer | Code-based design | Import design from existing code | Design system from codebase |
/workflow:ui-design:design-sync |
Designer | Syncing design | Synchronize design references | Updated design system docs |
/workflow:ui-design:reference-page-generator |
Developer | Reference docs | Generate component reference pages | Multi-component reference pages |
Memory
/memory:prepare
| Field | Description |
|---|---|
| Who | Developer preparing memory capture |
| When | Before capturing session knowledge |
| What | Delegates to universal-executor agent to prepare memory. Example: /memory:prepare "auth session" |
| How | Context gathering -> Memory structure preparation |
| Output | Prepared memory structure |
| Done | Ready for memory capture |
| Failure & Recovery | Agent failure handled |
| Impact & Boundary | Memory preparation only |
/memory:style-skill-memory
| Field | Description |
|---|---|
| Who | Developer documenting style knowledge |
| When | After learning project patterns |
| What | Generates SKILL memory package from style patterns. Example: /memory:style-skill-memory "React patterns" |
| How | Pattern extraction -> Skill memory generation |
| Output | SKILL memory package |
| Done | Skill memory documented |
| Failure & Recovery | Partial patterns preserved |
| Impact & Boundary | Style-specific memory |
CLI
/cli:codex-review
| Field | Description |
|---|---|
| Who | Developer reviewing code |
| When | Code review needed |
| What | Interactive code review using Codex CLI. Example: /cli:codex-review --uncommitted |
| How | Codex CLI -> Review analysis -> Interactive feedback |
| Output | Code review report |
| Done | Review complete with recommendations |
| Failure & Recovery | Codex unavailability fallback |
| Impact & Boundary | Git-aware code review |
/cli:cli-init
| Field | Description |
|---|---|
| Who | Developer setting up CLI config |
| When | First time setup |
| What | Generates .gemini/ and .qwen/ config directories. Example: /cli:cli-init |
| How | Create directories -> Generate default configs |
| Output | CLI configuration directories |
| Done | Configs ready for customization |
| Failure & Recovery | Existing configs preserved |
| Impact & Boundary | CLI configuration setup |
Section 2: Skills
Workflow Skills
workflow-plan
| Field | Description |
|---|---|
| Who | Developer planning tasks |
| When | Starting any planned work |
| What | Unified 4-phase planning: session discovery -> context gathering -> conflict resolution -> task generation |
| How | Phase 0: Session -> Phase 1: Context -> Phase 2: Conflicts -> Phase 3: Tasks |
| Output | plan.json + TASK-*.json in session folder |
| Done | Plan with TodoWrite tracking complete |
| Failure & Recovery | Conflict resolution prompts user |
| Impact & Boundary | Core planning skill |
workflow-execute
| Field | Description |
|---|---|
| Who | Developer executing planned tasks |
| When | After workflow-plan |
| What | Agent execution orchestration with lazy task loading, auto-commit |
| How | Load plan -> Spawn agents -> Execute -> Commit |
| Output | Code changes + task completion status |
| Done | All tasks executed or user-stopped |
| Failure & Recovery | Per-task retry, auto-rollback |
| Impact & Boundary | Core execution skill |
workflow-lite-plan
| Field | Description |
|---|---|
| Who | Developer needing quick planning |
| When | Simple tasks without full session |
| What | Lightweight planning - task analysis + breakdown |
| How | Analyze -> Break down -> Generate tasks |
| Output | Simplified plan.json |
| Done | Quick plan ready |
| Failure & Recovery | Falls back to workflow-plan for complex cases |
| Impact & Boundary | Lightweight alternative |
workflow-lite-execute
| Field | Description |
|---|---|
| Who | Developer executing simple plans |
| When | After lite-plan or standalone |
| What | Lightweight execution engine - multi-model collaboration |
| How | Load tasks -> Execute -> Verify |
| Output | Code changes |
| Done | Tasks completed |
| Failure & Recovery | Simple retry mechanism |
| Impact & Boundary | Lightweight alternative |
workflow-tdd-plan
| Field | Description |
|---|---|
| Who | Developer using TDD approach |
| When | Test-driven development |
| What | TDD workflow combining 6-phase planning with test-first |
| How | Plan tests -> Write tests -> Implement -> Verify |
| Output | Tests + implementation |
| Done | All tests passing |
| Failure & Recovery | Test failure iteration |
| Impact & Boundary | TDD-specific workflow |
workflow-test-fix
| Field | Description |
|---|---|
| Who | Developer fixing test failures |
| When | Tests failing |
| What | Test-fix pipeline combining test runner with fix agent |
| How | Run tests -> Identify failures -> Fix -> Re-run |
| Output | Passing tests |
| Done | All tests pass |
| Failure & Recovery | Iterative fix loop |
| Impact & Boundary | Test fixing workflow |
workflow-multi-cli-plan
| Field | Description |
|---|---|
| Who | Developer using multiple CLIs |
| When | Complex multi-perspective planning |
| What | Multi-CLI collaborative planning with ACE integration |
| How | Parallel CLI analysis -> Merge perspectives -> Generate plan |
| Output | Comprehensive plan from multiple models |
| Done | Merged plan complete |
| Failure & Recovery | Partial results merged |
| Impact & Boundary | Multi-model planning |
workflow-skill-designer
| Field | Description |
|---|---|
| Who | Developer creating skills |
| When | Building new workflow skills |
| What | Meta-skill for designing orchestrator+phase skills |
| How | Design -> Template -> Generate skill files |
| Output | New skill SKILL.md + phases |
| Done | Skill ready for use |
| Failure & Recovery | Validation errors guide fixes |
| Impact & Boundary | Skill meta-creation |
Team Skills
team-coordinate
| Field | Description |
|---|---|
| Who | Developer coordinating multi-agent work |
| When | Complex multi-role tasks |
| What | Universal team coordination with dynamic role generation |
| How | Analyze task -> Generate role-specs -> Dispatch workers -> Execute -> Deliver |
| Output | Coordinated multi-agent results |
| Done | Pipeline complete, artifacts delivered |
| Failure & Recovery | Role regeneration on capability gaps |
| Impact & Boundary | Universal team orchestrator |
team-edict
| Field | Description |
|---|---|
| Who | Developer using Edict architecture |
| When | Large-scale multi-agent coordination |
| What | Three-Six Ministry multi-agent framework replicating Edict architecture |
| How | Ministry-based role assignment -> Task dispatch -> Report aggregation |
| Output | Comprehensive multi-role deliverables |
| Done | All ministries report complete |
| Failure & Recovery | Ministry-level retry |
| Impact & Boundary | Large-scale coordination |
team-planex
| Field | Description |
|---|---|
| Who | Developer planning and executing |
| When | Plan-and-execute workflow |
| What | Unified team skill for plan-and-execute team |
| How | Plan -> Execute -> Verify |
| Output | Planned and executed work |
| Done | Plan executed completely |
| Failure & Recovery | Plan adjustment on failures |
| Impact & Boundary | Plan-execute team |
team-frontend
| Field | Description |
|---|---|
| Who | Frontend developer |
| When | Frontend development work |
| What | Unified team skill for frontend development |
| How | Component design -> Implementation -> Testing |
| Output | Frontend components and features |
| Done | Frontend work complete |
| Failure & Recovery | Component-level retry |
| Impact & Boundary | Frontend-specific team |
team-frontend-debug
| Field | Description |
|---|---|
| Who | Frontend developer debugging |
| When | Frontend issues |
| What | Frontend debugging team using Chrome DevTools MCP |
| How | Inspect -> Diagnose -> Fix -> Verify |
| Output | Fixed frontend issues |
| Done | Issues resolved |
| Failure & Recovery | Iterative debugging |
| Impact & Boundary | Frontend debugging specific |
team-testing
| Field | Description |
|---|---|
| Who | Developer writing tests |
| When | Test generation needed |
| What | Unified team skill for testing team |
| How | Test planning -> Generation -> Execution |
| Output | Test suites |
| Done | Tests passing |
| Failure & Recovery | Test iteration |
| Impact & Boundary | Testing-specific team |
team-review
| Field | Description |
|---|---|
| Who | Developer reviewing code |
| When | Code review needed |
| What | Unified team skill for code review |
| How | Multi-perspective review -> Feedback aggregation |
| Output | Review reports |
| Done | Review complete |
| Failure & Recovery | Partial review preserved |
| Impact & Boundary | Review-specific team |
team-quality-assurance
| Field | Description |
|---|---|
| Who | QA engineer |
| When | Quality verification |
| What | Unified team skill for quality assurance |
| How | Quality checks -> Issue identification -> Remediation |
| Output | QA reports and fixes |
| Done | Quality verified |
| Failure & Recovery | Issue-driven iteration |
| Impact & Boundary | QA-specific team |
team-arch-opt
| Field | Description |
|---|---|
| Who | Developer optimizing architecture |
| When | Architecture improvement |
| What | Unified team skill for architecture optimization |
| How | Analysis -> Optimization plan -> Implementation |
| Output | Improved architecture |
| Done | Optimization complete |
| Failure & Recovery | Gradual optimization |
| Impact & Boundary | Architecture-specific team |
team-perf-opt
| Field | Description |
|---|---|
| Who | Developer optimizing performance |
| When | Performance issues |
| What | Unified team skill for performance optimization |
| How | Profiling -> Bottleneck identification -> Optimization |
| Output | Performance improvements |
| Done | Performance targets met |
| Failure & Recovery | Iterative optimization |
| Impact & Boundary | Performance-specific team |
team-tech-debt
| Field | Description |
|---|---|
| Who | Developer addressing tech debt |
| When | Tech debt cleanup |
| What | Unified team skill for tech debt identification |
| How | Detection -> Prioritization -> Resolution |
| Output | Tech debt reduction |
| Done | Debt items resolved |
| Failure & Recovery | Prioritized resolution |
| Impact & Boundary | Tech debt specific |
team-ux-improve
| Field | Description |
|---|---|
| Who | UX designer/developer |
| When | UX improvements |
| What | Unified team skill for UX improvement |
| How | UX audit -> Improvement plan -> Implementation |
| Output | UX improvements |
| Done | UX enhanced |
| Failure & Recovery | User feedback iteration |
| Impact & Boundary | UX-specific team |
team-uidesign
| Field | Description |
|---|---|
| Who | UI designer |
| When | UI design work |
| What | Unified team skill for UI design team |
| How | Design system -> Components -> Prototypes |
| Output | UI designs and prototypes |
| Done | Design complete |
| Failure & Recovery | Design iteration |
| Impact & Boundary | UI design specific |
team-brainstorm
| Field | Description |
|---|---|
| Who | Team brainstorming |
| When | Ideation sessions |
| What | Unified team skill for brainstorming team |
| How | Multi-perspective ideation -> Synthesis |
| Output | Brainstorm results |
| Done | Ideas documented |
| Failure & Recovery | Partial synthesis |
| Impact & Boundary | Brainstorming specific |
team-issue
| Field | Description |
|---|---|
| Who | Developer managing issues |
| When | Issue resolution workflow |
| What | Unified team skill for issue resolution |
| How | Issue analysis -> Resolution -> Verification |
| Output | Resolved issues |
| Done | Issues closed |
| Failure & Recovery | Issue-specific handling |
| Impact & Boundary | Issue-specific team |
team-iterdev
| Field | Description |
|---|---|
| Who | Developer in iterative mode |
| When | Iterative development |
| What | Unified team skill for iterative development |
| How | Iteration -> Feedback -> Refinement |
| Output | Iteratively improved code |
| Done | Iteration goals met |
| Failure & Recovery | Iteration adjustment |
| Impact & Boundary | Iterative development team |
team-roadmap-dev
| Field | Description |
|---|---|
| Who | Developer on roadmap features |
| When | Roadmap-driven development |
| What | Unified team skill for roadmap-driven development |
| How | Roadmap parsing -> Feature development -> Tracking |
| Output | Roadmap feature completions |
| Done | Roadmap items complete |
| Failure & Recovery | Roadmap adjustment |
| Impact & Boundary | Roadmap-specific team |
team-ultra-analyze
| Field | Description |
|---|---|
| Who | Developer needing deep analysis |
| When | Complex analysis required |
| What | Deep collaborative analysis team skill |
| How | Multi-angle analysis -> Deep dive -> Synthesis |
| Output | Comprehensive analysis reports |
| Done | Analysis complete |
| Failure & Recovery | Partial analysis preserved |
| Impact & Boundary | Deep analysis team |
team-executor
| Field | Description |
|---|---|
| Who | Developer executing tasks |
| When | Task execution needed |
| What | Lightweight session execution skill |
| How | Load session -> Execute tasks -> Report |
| Output | Task execution results |
| Done | Session tasks complete |
| Failure & Recovery | Task-level retry |
| Impact & Boundary | Lightweight execution |
team-lifecycle-v4
| Field | Description |
|---|---|
| Who | Developer managing full lifecycle |
| When | Complete feature lifecycle |
| What | Full lifecycle team skill with clean architecture |
| How | Plan -> Develop -> Test -> Deploy -> Monitor |
| Output | Complete feature lifecycle |
| Done | Feature fully delivered |
| Failure & Recovery | Lifecycle stage retry |
| Impact & Boundary | Full lifecycle management |
team-designer
| Field | Description |
|---|---|
| Who | Developer designing teams |
| When | Creating new team skills |
| What | Meta-skill for generating team skills for different domains |
| How | Domain analysis -> Role definition -> Skill generation |
| Output | New team skill files |
| Done | Team skill ready |
| Failure & Recovery | Template validation |
| Impact & Boundary | Team skill meta-creation |
Utility Skills
brainstorm
| Field | Description |
|---|---|
| Who | Developer exploring ideas |
| When | Starting new work, exploring solutions |
| What | Unified brainstorming with dual-mode (auto/single-role) and 9 available roles |
| How | Role selection -> Perspective analysis -> Idea synthesis |
| Output | Brainstorm report with actionables |
| Done | Ideas documented |
| Failure & Recovery | Partial analysis preserved |
| Impact & Boundary | Ideation support |
spec-generator
| Field | Description |
|---|---|
| Who | Developer creating specifications |
| When | Starting new project or feature |
| What | 6-phase specification chain producing product brief, PRD, architecture, epics |
| How | Phase 1-6 chain: Brief -> PRD -> Architecture -> Epics -> Docs |
| Output | Complete specification documents |
| Done | All spec documents generated |
| Failure & Recovery | Phase-level retry |
| Impact & Boundary | Specification generation |
issue-manage
| Field | Description |
|---|---|
| Who | Developer managing issues |
| When | Issue CRUD operations |
| What | Interactive menu-driven interface for issue CRUD via ccw issue CLI |
| How | Menu selection -> CLI execution -> Result display |
| Output | Issue management operations |
| Done | Operation complete |
| Failure & Recovery | CLI error handling |
| Impact & Boundary | Issue CRUD only |
review-code
| Field | Description |
|---|---|
| Who | Developer reviewing code |
| When | Code review needed |
| What | Multi-dimensional code review with 6 dimensions: correctness, readability, performance, security, testing, architecture |
| How | Collect context -> Quick scan -> Deep review -> Generate report |
| Output | Structured review report |
| Done | Review complete |
| Failure & Recovery | Partial review preserved |
| Impact & Boundary | Code review only |
review-cycle
| Field | Description |
|---|---|
| Who | Developer in review cycle |
| When | Iterative code review |
| What | Unified multi-dimensional code review with iterative cycle |
| How | Review -> Feedback -> Fix -> Re-review |
| Output | Iteratively improved code |
| Done | All issues addressed |
| Failure & Recovery | Cycle iteration |
| Impact & Boundary | Iterative review |
memory-capture
| Field | Description |
|---|---|
| Who | Developer capturing knowledge |
| When | After important learnings |
| What | Unified memory capture with routing - session/tech/module context |
| How | Context detection -> Memory extraction -> Storage |
| Output | Captured memory entries |
| Done | Memory captured |
| Failure & Recovery | Partial capture preserved |
| Impact & Boundary | Memory capture only |
memory-manage
| Field | Description |
|---|---|
| Who | Developer managing memory |
| When | Memory maintenance |
| What | Unified memory management - CLAUDE.md updates and memory cleanup |
| How | Load memory -> Organize -> Update CLAUDE.md |
| Output | Updated memory files |
| Done | Memory organized |
| Failure & Recovery | Backup before changes |
| Impact & Boundary | Memory management |
skill-tuning
| Field | Description |
|---|---|
| Who | Developer improving skills |
| When | Skill optimization needed |
| What | Universal skill diagnosis and optimization |
| How | Diagnose -> Optimize -> Validate |
| Output | Improved skill files |
| Done | Skill optimized |
| Failure & Recovery | Incremental improvement |
| Impact & Boundary | Skill optimization |
skill-simplify
| Field | Description |
|---|---|
| Who | Developer simplifying skills |
| When | Skill too complex |
| What | SKILL.md simplification with functional preservation |
| How | Analyze -> Simplify -> Verify functionality |
| Output | Simplified skill file |
| Done | Skill simplified |
| Failure & Recovery | Functional verification |
| Impact & Boundary | Skill simplification |
skill-generator
| Field | Description |
|---|---|
| Who | Developer creating skills |
| When | New skill needed |
| What | Meta-skill for creating new Claude Code skills |
| How | Requirements -> Template -> Skill generation |
| Output | New skill SKILL.md + supporting files |
| Done | Skill ready for use |
| Failure & Recovery | Template validation |
| Impact & Boundary | Skill creation |
command-generator
| Field | Description |
|---|---|
| Who | Developer creating commands |
| When | New slash command needed |
| What | Command file generator - 5 phase workflow |
| How | Requirements -> Template -> Command generation |
| Output | New command .md file |
| Done | Command ready for use |
| Failure & Recovery | Schema validation |
| Impact & Boundary | Command creation |
ccw-help
| Field | Description |
|---|---|
| Who | Developer seeking help |
| When | Learning CCW system |
| What | CCW command help system - search, browse, documentation |
| How | Query -> Search -> Display help |
| Output | Help information |
| Done | Help displayed |
| Failure & Recovery | Fuzzy matching |
| Impact & Boundary | Help system only |
Appendix A: Quick Reference Table
Commands by Category
| Category | Commands | Count |
|---|---|---|
| Core | ccw, ccw-coordinator, flow-create | 3 |
| DDD | scan, index-build, plan, execute, doc-generate, doc-refresh, sync, update, auto | 9 |
| Issue | new, discover, discover-by-prompt, from-brainstorm, plan, queue, execute, convert-to-plan | 8 |
| IDAW | status, add, run, run-coordinate, resume | 5 |
| Workflow Session | start, list, sync, resume, complete | 5 |
| Workflow Spec | setup, add | 2 |
| Workflow Core | clean, debug-with-file, analyze-with-file, brainstorm-with-file, collaborative-plan-with-file | 5 |
| Workflow Cycle | integration-test-cycle, refactor-cycle, roadmap-with-file, unified-execute-with-file | 4 |
| Workflow UI Design | explore-auto, imitate-auto, generate, style-extract, layout-extract, animation-extract, codify-style, import-from-code, design-sync, reference-page-generator | 10 |
| Memory | prepare, style-skill-memory | 2 |
| CLI | codex-review, cli-init | 2 |
| Total Commands | 55 |
Skills by Category
| Category | Skills | Count |
|---|---|---|
| Workflow | workflow-plan, workflow-execute, workflow-lite-plan, workflow-lite-execute, workflow-tdd-plan, workflow-test-fix, workflow-multi-cli-plan, workflow-skill-designer | 8 |
| Team | team-coordinate, team-edict, team-planex, team-frontend, team-frontend-debug, team-testing, team-review, team-quality-assurance, team-arch-opt, team-perf-opt, team-tech-debt, team-ux-improve, team-uidesign, team-brainstorm, team-issue, team-iterdev, team-roadmap-dev, team-ultra-analyze, team-executor, team-lifecycle-v4, team-designer | 21 |
| Utility | brainstorm, spec-generator, issue-manage, review-code, review-cycle, memory-capture, memory-manage, skill-tuning, skill-simplify, skill-generator, command-generator, ccw-help | 12 |
| Total Skills | 41 |
Appendix B: Command Chain Examples
Document-Driven Development Flow
/ddd:scan # Code-first: scan existing codebase
-> /ddd:plan "add feature" # Plan with doc context
-> /ddd:execute # Execute with doc awareness
-> /ddd:sync # Sync changes to docs
Spec-First Development Flow
Skill: spec-generator # Generate specifications
-> /ddd:index-build # Build index from specs
-> /ddd:plan # Plan with spec context
-> /ddd:execute # Execute
-> /ddd:doc-refresh # Refresh affected docs
Issue Resolution Flow
/issue:new "bug description" # Create issue
-> /issue:plan ISS-001 # Generate solution
-> /issue:queue ISS-001 # Add to queue
-> /issue:execute # Execute queue
Team Coordination Flow
Skill: team-coordinate "implement auth"
-> Dynamic role generation
-> Parallel worker execution
-> Result aggregation
-> Completion action
UI Design Flow
/workflow:ui-design:explore-auto --input "modern dashboard"
-> /workflow:ui-design:style-extract # Extract styles
-> /workflow:ui-design:layout-extract # Extract layouts
-> /workflow:ui-design:generate # Assemble prototypes
Integration Test Flow
/workflow:integration-test-cycle "test auth flow"
-> Generate tests
-> Execute tests
-> Reflect on failures
-> Iterate until pass
-> Reflection log
Refactoring Flow
/workflow:refactor-cycle "src/auth"
-> Debt discovery
-> Prioritization
-> Per-item refactor
-> Validation
-> Reflection
Summary
| Metric | Count |
|---|---|
| Total Commands | 55 |
| Total Skills | 41 |
| DDD Commands | 9 |
| Issue Commands | 8 |
| Team Skills | 21 |
| Workflow Skills | 8 |
End of Reference Manual