Files
Claude-Code-Workflow/docs/skills/claude-collaboration.md
catlog22 1bf9006d65 Refactor Chinese documentation for team skills and commands
- Removed outdated table of contents from commands-skills.md
- Updated skills overview in claude-collaboration.md with new skill names and descriptions
- Enhanced clarity and structure of skills details, including roles and pipelines
- Added new team skills: team-arch-opt, team-perf-opt, team-brainstorm, team-frontend, team-uidesign, team-issue, team-iterdev, team-quality-assurance, team-roadmap-dev, team-tech-debt, team-ultra-analyze
- Improved user command section for better usability
- Streamlined best practices for team skills usage
2026-03-02 22:49:52 +08:00

517 lines
16 KiB
Markdown

# Claude Skills - Team Collaboration
## One-Line Positioning
**Team Collaboration Skills is a multi-role collaborative work orchestration system** — Through coordinator, worker roles, and message bus, it enables parallel processing and state synchronization for complex tasks.
## Pain Points Solved
| Pain Point | Current State | Claude Code Workflow Solution |
|------------|---------------|----------------------|
| **Single model limitation** | Can only call one AI model | Multi-role parallel collaboration, leveraging respective strengths |
| **Chaotic task orchestration** | Manual task dependency and state management | Automatic task discovery, dependency resolution, pipeline orchestration |
| **Fragmented collaboration** | Team members work independently | Unified message bus, shared state, progress sync |
| **Resource waste** | Repeated context loading | Wisdom accumulation, exploration cache, artifact reuse |
---
## Skills Overview
| Skill | Function | Use Case |
| --- | --- | --- |
| `team-coordinate-v2` | Universal team coordinator (dynamic role generation) | Any complex task |
| `team-lifecycle-v5` | Full lifecycle team (spec→impl→test) | Complete feature development |
| `team-planex` | Plan-execute pipeline | Issue batch processing |
| `team-review` | Code review team | Code review, vulnerability scanning |
| `team-testing` | Testing team | Test coverage, test case generation |
| `team-arch-opt` | Architecture optimization team | Refactoring, architecture analysis |
| `team-perf-opt` | Performance optimization team | Performance tuning, bottleneck analysis |
| `team-brainstorm` | Brainstorming team | Multi-angle analysis, idea generation |
| `team-frontend` | Frontend development team | UI development, design system |
| `team-uidesign` | UI design team | Design system, component specs |
| `team-issue` | Issue processing team | Issue analysis, implementation |
| `team-iterdev` | Iterative development team | Incremental delivery, agile development |
| `team-quality-assurance` | Quality assurance team | Quality scanning, defect management |
| `team-roadmap-dev` | Roadmap development team | Phased development, milestones |
| `team-tech-debt` | Tech debt team | Debt cleanup, code governance |
| `team-ultra-analyze` | Deep analysis team | Complex problem analysis, collaborative exploration |
| `team-executor-v2` | Lightweight executor | Session resume, pure execution |
---
## Core Architecture
All Team Skills share a unified **team-worker agent architecture**:
```
┌──────────────────────────────────────────────────────────┐
│ Skill(skill="team-xxx", args="task description") │
└────────────────────────┬─────────────────────────────────┘
│ Role Router
┌──── --role present? ────┐
│ NO │ YES
↓ ↓
Orchestration Mode Role Dispatch
(auto → coordinator) (route to role.md)
┌─────────┴─────────┬───────────────┬──────────────┐
↓ ↓ ↓ ↓
┌────────┐ ┌────────┐ ┌────────┐ ┌────────┐
│ coord │ │worker 1│ │worker 2│ │worker N│
│(orchestrate)│ │(execute)│ │(execute)│ │(execute)│
└────────┘ └────────┘ └────────┘ └────────┘
│ │ │ │
└───────────────────┴───────────────┴──────────────┘
Message Bus (message bus)
```
**Core Components**:
- **Coordinator**: Built-in orchestrator for task analysis, dispatch, monitoring
- **Team-Worker Agent**: Unified agent, loads role-spec to execute role logic
- **Role Router**: `--role=xxx` parameter routes to role execution
- **Message Bus**: Inter-team member communication protocol
- **Shared Memory**: Cross-task knowledge accumulation (Wisdom)
---
## Skills Details
### team-coordinate-v2
**One-Liner**: Universal team coordinator — Dynamically generates roles and orchestrates execution based on task analysis
**Trigger**:
```bash
team-coordinate <task-description>
team-coordinate --role=coordinator <task>
```
**Features**:
- Only coordinator is built-in, all worker roles are dynamically generated at runtime
- Supports inner-loop roles (handle multiple same-prefix tasks)
- Fast-Advance mechanism skips coordinator to directly spawn successor tasks
- Wisdom accumulates cross-task knowledge
**Session Directory**:
```
.workflow/.team/TC-<slug>-<date>/
├── team-session.json # Session state + dynamic role registry
├── task-analysis.json # Phase 1 output
├── roles/ # Dynamic role definitions
├── artifacts/ # All MD deliverables
├── wisdom/ # Cross-task knowledge
└── .msg/ # Team message bus logs
```
---
### team-lifecycle-v5
**One-Liner**: Full lifecycle team — Complete pipeline from specification to implementation to testing to review
**Trigger**:
```bash
team-lifecycle <task-description>
```
**Features**:
- Based on team-worker agent architecture, all workers share the same agent definition
- Role-specific Phase 2-4 loaded from markdown specs
- Supports specification pipeline, implementation pipeline, frontend pipeline
**Role Registry**:
| Role | Spec | Task Prefix | Inner Loop |
|------|------|-------------|------------|
| coordinator | roles/coordinator/role.md | (none) | - |
| analyst | role-specs/analyst.md | RESEARCH-* | false |
| writer | role-specs/writer.md | DRAFT-* | true |
| planner | role-specs/planner.md | PLAN-* | true |
| executor | role-specs/executor.md | IMPL-* | true |
| tester | role-specs/tester.md | TEST-* | false |
| reviewer | role-specs/reviewer.md | REVIEW-* | false |
**Pipeline Definitions**:
```
Specification Pipeline: RESEARCH → DRAFT → QUALITY
Implementation Pipeline: PLAN → IMPL → TEST + REVIEW
Full Lifecycle: [Spec Pipeline] → [Impl Pipeline]
```
---
### team-planex
**One-Liner**: Plan-and-execute team — Per-issue beat pipeline
**Trigger**:
```bash
team-planex <task-description>
team-planex --role=planner <input>
team-planex --role=executor --input <solution-file>
```
**Features**:
- 2-member team (planner + executor), planner serves as lead role
- Per-issue beat: planner creates EXEC-* task immediately after completing each issue
- Solution written to intermediate artifact file, executor loads from file
**Wave Pipeline**:
```
Issue 1: planner plans → write artifact → create EXEC-* → executor executes
Issue 2: planner plans → write artifact → create EXEC-* → executor parallel consume
Final: planner sends all_planned → executor completes remaining → finish
```
---
### team-review
**One-Liner**: Code review team — Unified code scanning, vulnerability review, auto-fix
**Trigger**:
```bash
team-review <target-path>
team-review --full <target-path> # scan + review + fix
team-review --fix <review-files> # fix only
team-review -q <target-path> # quick scan only
```
**Role Registry**:
| Role | Task Prefix | Type |
|------|-------------|------|
| coordinator | RC-* | orchestrator |
| scanner | SCAN-* | read-only analysis |
| reviewer | REV-* | read-only analysis |
| fixer | FIX-* | code generation |
**Pipeline**:
```
SCAN-* (scan) → REV-* (review) → [User confirmation] → FIX-* (fix)
```
**Review Dimensions**: Security, Correctness, Performance, Maintainability
---
### team-testing
**One-Liner**: Testing team — Progressive test coverage through Generator-Critic loop
**Trigger**:
```bash
team-testing <task-description>
```
**Role Registry**:
| Role | Task Prefix | Type |
|------|-------------|------|
| coordinator | (none) | orchestrator |
| strategist | STRATEGY-* | pipeline |
| generator | TESTGEN-* | pipeline |
| executor | TESTRUN-* | pipeline |
| analyst | TESTANA-* | pipeline |
**Three Pipelines**:
```
Targeted: STRATEGY → TESTGEN(L1) → TESTRUN
Standard: STRATEGY → TESTGEN(L1) → TESTRUN → TESTGEN(L2) → TESTRUN → TESTANA
Comprehensive: STRATEGY → [TESTGEN(L1+L2) parallel] → [TESTRUN parallel] → TESTGEN(L3) → TESTRUN → TESTANA
```
**Test Layers**: L1: Unit (80%) → L2: Integration (60%) → L3: E2E (40%)
---
### team-arch-opt
**One-Liner**: Architecture optimization team — Analyze architecture issues, design refactoring strategies, implement improvements
**Trigger**:
```bash
team-arch-opt <task-description>
```
**Role Registry**:
| Role | Task Prefix | Function |
|------|-------------|----------|
| coordinator | (none) | orchestrator |
| analyzer | ANALYZE-* | architecture analysis |
| designer | DESIGN-* | refactoring design |
| refactorer | REFACT-* | implement refactoring |
| validator | VALID-* | validate improvements |
| reviewer | REVIEW-* | code review |
**Detection Scope**: Dependency cycles, coupling/cohesion, layering violations, God Classes, dead code
---
### team-perf-opt
**One-Liner**: Performance optimization team — Performance profiling, bottleneck identification, optimization implementation
**Trigger**:
```bash
team-perf-opt <task-description>
```
**Role Registry**:
| Role | Task Prefix | Function |
|------|-------------|----------|
| coordinator | (none) | orchestrator |
| profiler | PROFILE-* | performance profiling |
| strategist | STRAT-* | optimization strategy |
| optimizer | OPT-* | implement optimization |
| benchmarker | BENCH-* | benchmarking |
| reviewer | REVIEW-* | code review |
---
### team-brainstorm
**One-Liner**: Brainstorming team — Multi-angle creative analysis, Generator-Critic loop
**Trigger**:
```bash
team-brainstorm <topic>
team-brainstorm --role=ideator <topic>
```
**Role Registry**:
| Role | Task Prefix | Function |
|------|-------------|----------|
| coordinator | (none) | orchestrator |
| ideator | IDEA-* | idea generation |
| challenger | CHALLENGE-* | critical questioning |
| synthesizer | SYNTH-* | synthesis integration |
| evaluator | EVAL-* | evaluation scoring |
---
### team-frontend
**One-Liner**: Frontend development team — Built-in ui-ux-pro-max design intelligence
**Trigger**:
```bash
team-frontend <task-description>
```
**Role Registry**:
| Role | Task Prefix | Function |
|------|-------------|----------|
| coordinator | (none) | orchestrator |
| analyst | ANALYZE-* | requirement analysis |
| architect | ARCH-* | architecture design |
| developer | DEV-* | frontend implementation |
| qa | QA-* | quality assurance |
---
### team-uidesign
**One-Liner**: UI design team — Design system analysis, token definition, component specs
**Trigger**:
```bash
team-uidesign <task>
```
**Role Registry**:
| Role | Task Prefix | Function |
|------|-------------|----------|
| coordinator | (none) | orchestrator |
| researcher | RESEARCH-* | design research |
| designer | DESIGN-* | design definition |
| reviewer | AUDIT-* | accessibility audit |
| implementer | BUILD-* | code implementation |
---
### team-issue
**One-Liner**: Issue processing team — Issue processing pipeline
**Trigger**:
```bash
team-issue <issue-ids>
```
**Role Registry**:
| Role | Task Prefix | Function |
|------|-------------|----------|
| coordinator | (none) | orchestrator |
| explorer | EXPLORE-* | code exploration |
| planner | PLAN-* | solution planning |
| implementer | IMPL-* | code implementation |
| reviewer | REVIEW-* | code review |
| integrator | INTEG-* | integration validation |
---
### team-iterdev
**One-Liner**: Iterative development team — Generator-Critic loop, incremental delivery
**Trigger**:
```bash
team-iterdev <task-description>
```
**Role Registry**:
| Role | Task Prefix | Function |
|------|-------------|----------|
| coordinator | (none) | orchestrator |
| architect | ARCH-* | architecture design |
| developer | DEV-* | feature development |
| tester | TEST-* | test validation |
| reviewer | REVIEW-* | code review |
**Features**: Developer-Reviewer loop (max 3 rounds), Task Ledger real-time progress
---
### team-quality-assurance
**One-Liner**: Quality assurance team — Issue discovery + test validation closed loop
**Trigger**:
```bash
team-quality-assurance <task-description>
team-qa <task-description>
```
**Role Registry**:
| Role | Task Prefix | Function |
|------|-------------|----------|
| coordinator | (none) | orchestrator |
| scout | SCOUT-* | issue discovery |
| strategist | QASTRAT-* | strategy formulation |
| generator | QAGEN-* | test generation |
| executor | QARUN-* | test execution |
| analyst | QAANA-* | result analysis |
---
### team-roadmap-dev
**One-Liner**: Roadmap development team — Phased development, milestone management
**Trigger**:
```bash
team-roadmap-dev <task-description>
```
**Role Registry**:
| Role | Task Prefix | Function |
|------|-------------|----------|
| coordinator | (none) | human interaction |
| planner | PLAN-* | phase planning |
| executor | EXEC-* | phase execution |
| verifier | VERIFY-* | phase validation |
---
### team-tech-debt
**One-Liner**: Tech debt team — Debt scanning, assessment, cleanup, validation
**Trigger**:
```bash
team-tech-debt <task-description>
```
**Role Registry**:
| Role | Task Prefix | Function |
|------|-------------|----------|
| coordinator | (none) | orchestrator |
| scanner | TDSCAN-* | debt scanning |
| assessor | TDEVAL-* | quantitative assessment |
| planner | TDPLAN-* | governance planning |
| executor | TDFIX-* | cleanup execution |
| validator | TDVAL-* | validation regression |
---
### team-ultra-analyze
**One-Liner**: Deep analysis team — Multi-role collaborative exploration, progressive understanding
**Trigger**:
```bash
team-ultra-analyze <topic>
team-analyze <topic>
```
**Role Registry**:
| Role | Task Prefix | Function |
|------|-------------|----------|
| coordinator | (none) | orchestrator |
| explorer | EXPLORE-* | code exploration |
| analyst | ANALYZE-* | deep analysis |
| discussant | DISCUSS-* | discussion interaction |
| synthesizer | SYNTH-* | synthesis output |
**Features**: Supports Quick/Standard/Deep three depth modes
---
### team-executor-v2
**One-Liner**: Lightweight executor — Resume session, pure execution mode
**Trigger**:
```bash
team-executor --session=<path>
```
**Features**:
- No analysis, no role generation — only load and execute existing session
- Used to resume interrupted team-coordinate sessions
---
## User Commands
All Team Skills support unified user commands (wake paused coordinator):
| Command | Action |
|---------|--------|
| `check` / `status` | Output execution status graph, no progress |
| `resume` / `continue` | Check worker status, advance next step |
| `revise <TASK-ID>` | Create revision task + cascade downstream |
| `feedback <text>` | Analyze feedback impact, create targeted revision chain |
---
## Best Practices
1. **Choose the right team type**:
- General tasks → `team-coordinate-v2`
- Complete feature development → `team-lifecycle-v5`
- Issue batch processing → `team-planex`
- Code review → `team-review`
- Test coverage → `team-testing`
- Architecture optimization → `team-arch-opt`
- Performance tuning → `team-perf-opt`
- Brainstorming → `team-brainstorm`
- Frontend development → `team-frontend`
- UI design → `team-uidesign`
- Tech debt → `team-tech-debt`
- Deep analysis → `team-ultra-analyze`
2. **Leverage inner-loop roles**: Set `inner_loop: true` to let single worker handle multiple same-prefix tasks
3. **Wisdom accumulation**: All roles in team sessions accumulate knowledge to `wisdom/` directory
4. **Fast-Advance**: Simple linear successor tasks automatically skip coordinator to spawn directly
5. **Checkpoint recovery**: All team skills support session recovery via `--resume` or `resume` command
---
## Related Commands
- [Claude Commands - Workflow](../commands/claude/workflow.md)
- [Claude Commands - Session](../commands/claude/session.md)