Agent skill

workflow-orchestrator

Orchestrate the six-stage development workflow. Use when user asks "what stage am I at?", "what's next?", "workflow status", "where should I start?", "工作流程", "下一步是什麼", "我在哪個階段", or wants guidance on the development process. Detects current state by checking changes/ folder and guides users through brainstorm → spec → plan → tdd → review → archive stages.

Stars 163
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/workflow-orchestrator

SKILL.md

Workflow Orchestrator

🎯 Purpose: Detect current workflow stage and guide users through the six-stage development process with clear Agent recommendations.

When to Use This Skill

Use this skill when:

  • Starting a new feature and unsure where to begin
  • Wondering "what's next?" after completing a stage
  • Need to check current workflow status
  • Want to understand the overall development flow
  • 想知道目前在哪個階段
  • 不確定下一步該做什麼

Six-Stage Workflow Overview

1. Brainstorm → 2. Spec → 3. Plan → 4. TDD → 5. Review → 6. Archive

Standard Path (Med/High risk):

Brainstorm → Spec → Plan → TDD → Review → Archive

Fast Path (Low risk only):

Brainstorm → Plan → TDD → Review → Archive
          (skip Spec)

Current State Detection

Step 1: Check Changes Folder

Look for changes/<YYYY-MM-DD>-<slug>/ directory and check which files exist:

File Present Stage Completed Next Stage Recommended Action
None Not started Brainstorm Start with brainstorming
01-brainstorm.md Brainstorm ✅ Spec Generate specification
03-spec.md Spec ✅ Plan Create implementation plan
04-plan.md Plan ✅ TDD Start TDD implementation
Code changes (git diff) TDD in progress/done Review Run code review
05-review.md Review ✅ Archive Finalize and archive
99-archive.md Complete 🎉 - Start new feature

Step 2: Provide Guidance

Based on detected stage, provide:

  1. Current Status: Which stage is completed
  2. 🎯 Next Step: What to do next
  3. 🤖 Agent Recommendation: Which agent to use
  4. 📝 Action Instructions: Clear CLI/VS Code commands
  5. 🔗 Reference: Link to detailed documentation

Stage-by-Stage Guidance

Stage 0: Not Started → Brainstorm

Status: No change package detected

Next Step: Start brainstorming session

Recommended Agent: architect-agent or spec-agent

Action Instructions:

CLI:

Input: "我要開始一個新功能的 brainstorming"
[System loads brainstorming skill]
→ /agent → Select architect-agent or spec-agent
→ Continue conversation to generate 01-brainstorm.md

VS Code:

Input: /brainstorm
Or: "brainstorm a new feature"
→ Select @workspace #architect-agent

Triggers brainstorming skill: Risk classification, option exploration, decision log


Stage 1: Brainstorm ✅ → Spec

Status: 01-brainstorm.md exists, clarified requirements

Next Step: Generate specification document (PRD)

Recommended Agent: spec-agent

Action Instructions:

CLI:

Input: "產生 spec 文件"
[System loads specification skill]
→ /agent → Select spec-agent
→ Continue conversation to generate 03-spec.md

VS Code:

Input: /spec
Or: "generate spec document"
→ Select @workspace #spec-agent

Triggers specification skill: User stories, acceptance criteria, technical requirements

Note: Low-risk changes can skip to Plan (Fast Path)


Stage 2: Spec ✅ → Plan

Status: 03-spec.md exists, requirements documented

Next Step: Break down into executable implementation plan

Recommended Agent: plan-agent

Action Instructions:

CLI:

Input: "幫我規劃實作計畫"
[System loads implementation-planning skill]
→ /agent → Select plan-agent
→ Continue conversation to generate 04-plan.md

VS Code:

Input: /create-plan
Or: "create implementation plan"
→ Select @workspace #plan-agent

Triggers implementation-planning skill: Task breakdown, TDD integration, impact analysis


Stage 3: Plan ✅ → TDD

Status: 04-plan.md exists, tasks defined

Next Step: Start test-driven development implementation

Recommended Agent: coder-agent

Action Instructions:

CLI:

Input: "開始 TDD 實作"
[System loads tdd-workflow skill]
→ /agent → Select coder-agent
→ Follow Red-Green-Refactor cycle

VS Code:

Input: /tdd
Or: "start TDD implementation"
→ Select @workspace #coder-agent

Triggers tdd-workflow skill: Red-Green-Refactor, test scaffolding, coverage verification


Stage 4: TDD Done → Review

Status: Code changes detected (git status/diff shows modifications)

Next Step: Run code and security review

Recommended Agent: code-reviewer-agent

Action Instructions:

CLI:

Input: "review 我的 code"
[System loads code-security-review skill]
→ /agent → Select code-reviewer-agent
→ Review checklist: DDD, security, financial precision

VS Code:

Input: /code-review
Or: "review my code"
→ Select @workspace #code-reviewer-agent

Triggers code-security-review skill: Code quality, security audit, financial correctness


Stage 5: Review ✅ → Archive

Status: 05-review.md exists, code reviewed and approved

Next Step: Finalize change package and create work log

Recommended Agent: Default agent (no specific agent required)

Action Instructions:

CLI:

Input: "archive 這個 change package"
[System loads work-archiving skill]
→ Generate 99-archive.md and update WORK_LOG.md

VS Code:

Input: /archive
Or: "finalize and archive"

Triggers work-archiving skill: Work log entry, lessons learned, follow-up items


Stage 6: Complete 🎉

Status: 99-archive.md exists, change package archived

Next Step: Start new feature or celebrate! 🎉

Action: Ready for next work item

  • Run /workflow for new feature
  • Or: "我要開始新的 brainstorming"

Output Template

When user asks about workflow status, respond with:

markdown
## 📍 Current Workflow Status

### Detected State
- **Change Package**: `changes/<date>-<slug>/`
- **Completed Stages**: [List completed stages with ✅]
- **Current Stage**: [Current stage name]

### 🎯 Next Step: [Next stage name]

**What to do**: [Brief description]

**Recommended Agent**: `[agent-name]`

**Action Instructions**:

**CLI**:
[Provide exact CLI commands and natural language inputs]

**VS Code**:
[Provide slash command or natural language inputs]

### 📊 Progress Overview
- [x] Brainstorm
- [x] Spec
- [ ] Plan ← **You are here**
- [ ] TDD
- [ ] Review
- [ ] Archive

### 📖 Detailed Workflow
For complete workflow documentation, see [WORKFLOW.md](../../WORKFLOW.md)

Fast Path Decision Helper

If user asks "should I skip spec?", apply these rules:

Risk Level Criteria Workflow Path
Low Bug fix, minor refactor, config change Fast Path (skip Spec)
Medium New feature, API change, schema change Standard Path (include Spec)
High Security, auth, money handling, breaking change Standard Path (mandatory)

When in doubt: Use Standard Path (better safe than sorry)

Troubleshooting

"I don't see any changes/ folder"

Solution: You haven't started a change package yet. Begin with brainstorming:

  • CLI: "開始 brainstorming"
  • VS Code: /brainstorm

"Multiple change folders exist"

Solution: Work on the most recent one (highest date). Archive old ones if completed.

"I want to skip a stage"

Solution:

  • Low-risk only: Can skip Spec (brainstorm → plan)
  • Never skip: Brainstorm, Plan, TDD, Review (core safety net)

"Which agent should I use?"

Solution: Follow the "Recommended Agent" for each stage. If unsure, use the default agent—it will still work but may be less specialized.

Integration with Other Skills

This orchestrator skill works with:

  • brainstorming: Stage 1
  • specification: Stage 2
  • implementation-planning: Stage 3
  • tdd-workflow: Stage 4
  • code-security-review: Stage 5
  • work-archiving: Stage 6

Related Documentation

  • WORKFLOW.md - Detailed six-stage process
  • AGENTS.md - Agent roles and responsibilities
  • README.zh-TW.md - CLI vs VS Code usage guide

💡 Pro Tip: Run workflow detection regularly by asking "what's next?" to stay on track and maintain momentum.

Expand your agent's capabilities with these related and highly-rated skills.

Didn't find tool you were looking for?

Be as detailed as possible for better results