tim-hub/role-based-skills/coordinator
- community
Checking that the SKILL.md file is still in the repository.
Orchestrate multiple skills through a phased workflow to solve problems, generate ideas, create plans, or build software. Use when the user asks to coordinate roles, run a full workflow, or when a task benefits from brainstorming, challenging, planning, and implementing in sequence.
Install
npx vskill@latest install tim-hub/role-based-skills/coordinatorSKILL.md
Coordinator
Orchestrate skills through a phased workflow: brainstorm, challenge, design/plan, challenge, implement, finish.
Announce at start: "Using the coordinator skill to orchestrate this workflow."
Determine Scope
Classify the request before starting:
| Scope | Phases | Example | |
|---|---|---|---|
| Idea | 1-2 | "Help me think through X" | |
| Plan | 1-4 | "Plan how to build X" | |
| Build | 1-6 | "Build X for me" |
Ask if unclear: "Is this an idea exploration, a plan, or a full build?"
Phases
Phase 1: Brainstorm
- Load and apply the brainstormer skill
- Generate diverse ideas across multiple thinking frameworks
- Produce 2-4 ideas per framework, covering 3-5 frameworks
- Load and apply the brainstorming skill
- Refine the best ideas through collaborative dialogue
- Present design in sections, validate incrementally
Output: Refined ideas with rationale and top picks.
Checkpoint: Present top ideas. Continue unless redirected.
Phase 2: Challenge Ideas
- Load and apply the challenger skill
- Attack ideas across all dimensions (security, correctness, performance, scalability, UX, maintainability, edge cases)
Gate:
- Critical issues found → STOP. Present issues, ask how to proceed.
- Medium/low only → Summarize findings, continue automatically.
Output: Challenge report with severity ratings.
For Idea scope: Stop here. Present final deliverable.
Phase 3: Design & Plan
Run applicable sub-phases:
3a. Design (if UI/UX involved):
- Load and apply the designer skill
- Review and improve UX, produce design recommendations
3b. Plan:
- Load and apply the planner skill
- Create structured plan with approaches, trade-offs, risks
3c. Implementation Plan (Build scope only):
- Load and apply the writing-plans skill
- Create detailed task-level implementation plan with file paths and code examples
Checkpoint: Present the plan. Continue unless redirected.
For Plan scope: Stop here. Present final deliverable.
Phase 4: Challenge Plan
- Load and apply the challenger skill again
- Evaluate the plan for risks, missing steps, wrong ordering, feasibility
Gate:
- Critical issues found → Revise plan based on findings, re-challenge. Max 2 revision cycles.
- Medium/low only → Summarize, continue automatically.
Output: Validated, battle-tested plan.
Phase 5: Implement
- Load and apply the test-driven-development skill
- TDD is mandatory: no production code without a failing test first
- Load and apply the executing-plans skill
- Execute plan in controlled batches with review checkpoints
Within each batch:
- RED: Write failing test
- GREEN: Minimal code to pass
- REFACTOR: Clean up, stay green
- Review between batches
Gate:
- Tests failing after 2 fix attempts → STOP. Report issue, ask for direction.
- Batch complete → Brief summary, continue to next batch.
Phase 6: Finish
- Load and apply the finishing-a-development-branch skill
- Verify all tests pass
- Present completion options (merge, PR, keep, discard)
- Clean up worktree if applicable
Autonomy Rules
| Situation | Action | |
|---|---|---|
| Challenger finds critical issues | STOP — present issues, ask for direction | |
| Challenger finds medium/low issues | Summarize and continue | |
| Between major phases | Brief checkpoint, continue unless redirected | |
| Tests failing during implementation | Attempt fix twice; if stuck, STOP | |
| Ambiguous requirements discovered | STOP — ask for clarification | |
| Phase not applicable to request | Skip with brief note |
Workflow Diagram
Key Principles
- One skill at a time — Load and complete each skill before moving to the next.
- Challenge gates — Critical issues block progress; medium/low don't.
- TDD is non-negotiable — No implementation code without failing tests first.
- Incremental validation — Check with human at phase boundaries.
- Skip what doesn't apply — Not every request needs all phases.
- Autonomy with guardrails — Run freely, stop when uncertain or blocked.
Skills Referenced
| Skill | Phase | Purpose | |
|---|---|---|---|
| brainstormer | 1 | Generate diverse ideas via thinking frameworks | |
| brainstorming | 1 | Refine ideas through collaborative dialogue | |
| challenger | 2, 4 | Adversarial evaluation of ideas and plans | |
| designer | 3a | UX review and improvement | |
| planner | 3b | Structured plan with trade-offs | |
| writing-plans | 3c | Detailed implementation tasks | |
| executing-plans | 5 | Batch execution with review checkpoints | |
| test-driven-development | 5 | Red-green-refactor cycle | |
| finishing-a-development-branch | 6 | Branch completion and cleanup |
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingcoordinator/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/tim-hub/role-based-skills/coordinator)