Execute 9-phase workflow for complex features. DO NOT use for simple bug fixes.
/plugin marketplace add nguyenthienthanh/aura-frog/plugin install aura-frog@aurafrogThis skill is limited to using the following tools:
phase-skipping.mdPriority: CRITICAL - Use for complex feature implementations Version: 1.0.0
USE for:
DON'T use for:
bugfix-quick| Phase | Name | Lead Agent | Deliverable | Gate |
|---|---|---|---|---|
| 1 | Understand 🎯 | pm-operations-orchestrator | Requirements document | ✋ |
| 2 | Design 🏗️ | Dev agent | Technical design | ✋ |
| 3 | UI Breakdown 🎨 | ui-designer | Component breakdown | ✋ |
| 4 | Plan Tests 🧪 | qa-automation | Test strategy | ✋ |
| 5a | Write Tests 🔴 | qa-automation + Dev | Failing tests (TDD RED) | ✋ |
| 5b | Build 🟢 | Dev agent | Implementation (TDD GREEN) | ✋ |
| 5c | Polish ♻️ | Dev agent | Refactored code (TDD REFACTOR) | ✋ |
| 6 | Review 👀 | security-expert | Quality review report | ✋ |
| 7 | Verify ✅ | qa-automation | All tests pass, coverage ≥80% | ✋ |
| 8 | Document 📚 | pm-operations-orchestrator | Documentation | ✋ |
| 9 | Share 🔔 | slack-operations | Team notification | Auto |
Phase 1 (Understand) → Phase 2 (Design)
Requires: approve
Blocker: Unclear requirements
Phase 2 (Design) → Phase 3 (UI)
Requires: approve
Blocker: No technical design
Phase 3 (UI) → Phase 4 (Plan Tests)
Requires: approve
Skip if: No UI component in task
Phase 4 (Plan Tests) → Phase 5a (Write Tests)
Requires: approve
Blocker: No test strategy
Phase 5a (RED) → Phase 5b (GREEN)
Requires: approve + tests MUST fail
Blocker: Tests pass (they should fail!)
Phase 5b (GREEN) → Phase 5c (REFACTOR)
Requires: approve + tests MUST pass
Blocker: Tests still failing
Phase 5c (REFACTOR) → Phase 6 (Review)
Requires: approve + tests MUST still pass
Blocker: Tests broken by refactor
Phase 6 (Review) → Phase 7 (Verify)
Requires: approve
Blocker: Critical security issues
Phase 7 (Verify) → Phase 8 (Document)
Requires: approve + coverage ≥80%
Blocker: Coverage below target
Phase 8 (Document) → Phase 9 (Share)
Requires: approve
Auto-executes Phase 9
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🏗️ Phase [N]: [Name] - Approval Needed
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
## [Friendly Summary] ✨
[Deliverables list]
---
📍 **Progress:** ████░░░░░░░░░░░░ [X]% ([N]/8 phases)
⏭️ **After approval:**
→ Phase [N+1]: [Name] - [Brief description]
→ Will skip: [Phase X] ([reason]) — if applicable
---
**Options:**
- `approve` / `yes` → Continue to Phase [N+1]
- `reject: <reason>` → Brainstorm & re-do
- `modify: <changes>` → Adjust deliverables
- `stop` → Cancel workflow
⚡ After approval, I'll AUTO-CONTINUE to Phase [N+1]!
See: rules/workflow-navigation.md for full navigation format.
| Response | Action |
|---|---|
approve / yes | Continue to next phase immediately |
reject: <reason> | Brainstorm first, then restart phase |
modify: <changes> | Light brainstorm, then adjust deliverables |
stop / cancel | End workflow, save state |
Default behavior: Brainstorm feedback before implementing.
Force mode: Skip brainstorming when user says:
See: rules/feedback-brainstorming.md
After user approves:
workflow:handoffExample Flow:
User: "approve"
→ Claude immediately starts Phase 3
→ Completes Phase 3 deliverables
→ Shows Phase 3 approval gate
→ Waits for user
User: "approve"
→ Claude immediately starts Phase 4
→ ... continues ...
Token Awareness:
workflow:handoffPhase 5a (RED):
✅ Write tests FIRST
✅ Run tests → MUST FAIL
❌ If tests pass → STOP, tests aren't testing new code
Phase 5b (GREEN):
✅ Write minimal code to pass tests
✅ Run tests → MUST PASS
❌ If tests fail → Fix code, not tests
Phase 5c (REFACTOR):
✅ Clean up code
✅ Run tests → MUST STILL PASS
❌ If tests fail → Revert refactor
| Phase | Creator | Reviewers |
|---|---|---|
| 1 (Understand) | PM | Dev + QA + UI |
| 2 (Design) | Dev | Secondary Dev + QA |
| 4 (Plan Tests) | QA | Dev |
| 6 (Review) | Security | All |
User can request skip with reason:
User: "skip phase 3, this is backend only"
→ Log skip reason
→ Proceed to Phase 4
docs/phases/phase-1-understand.md
docs/phases/phase-2-design.md
docs/phases/phase-3-ui.md
docs/phases/phase-4-test-planning.md
docs/phases/phase-5-implementation.md
docs/phases/phase-6-review.md
docs/phases/phase-7-verification.md
docs/phases/phase-8-documentation.md
docs/phases/phase-9-notification.md
.claude/project-contexts/[project]/project-config.yaml
.claude/project-contexts/[project]/conventions.md
.claude/project-contexts/[project]/rules.md
rules/tdd-workflow.md
rules/kiss-principle.md
rules/code-quality.md
rules/feedback-brainstorming.md
rules/workflow-navigation.md
workflow:handoff
→ Saves to .claude/logs/workflows/[workflow-id]/
→ Contains: current phase, deliverables, context
workflow:resume <workflow-id>
→ Loads saved state
→ Continues from last phase
→ Re-shows approval gate if pending
workflow:status
→ Shows: current phase, completed phases, pending tasks
User: "workflow:start Add user authentication with JWT"
Phase 1: Understand 🎯
├── Analyze requirements
├── Define success criteria
├── Identify risks
├── Cross-review with dev/QA/UI
└── Show approval gate ✋
User: "approve"
Phase 2: Design 🏗️ [AUTO-CONTINUE]
├── Design JWT architecture
├── Define API endpoints
├── Plan database schema
├── Cross-review
└── Show approval gate ✋
User: "approve"
Phase 3: UI Breakdown 🎨 [AUTO-CONTINUE]
├── Design login/register screens
├── Extract design tokens
├── Plan component structure
└── Show approval gate ✋
... continues through all phases ...
Phase 9: Share 🔔 [AUTO-EXECUTE]
├── Send Slack notification
├── Update JIRA ticket
└── Workflow complete ✅
Remember:
rules/workflow-navigation.md)This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.
This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.