[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6DPX9ZX3EK4ZX3RP3502WW3":3,"skill-same:01M6DPX9ZX3EK4ZX3RP3502WW3":74},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":26,"publishedAt":46,"linkOnly":47,"markdownContent":48,"markdownBody":49,"frontMatter":50,"refName":51,"commitSha":52,"skillDirectory":53,"forksCount":54,"repositoryHtmlUrl":55,"repositoryHomepageUrl":6,"ownerHtmlUrl":56,"ownerBlogUrl":57,"ownerDescription":58,"updatedAt":59,"install":60,"resources":73},"01M6DPX9ZX3EK4ZX3RP3502WW3","rlm-subagent",null,"Master skill for parallel subagent-driven execution with automatic fallback to single-agent sequential mode. Use when implementing plans with multiple independent sub-phases (SP1, SP2...) to dispatch parallel subagents, or when requiring code review between implementation and testing. Trigger phrases: \"parallelize\", \"dispatch subagent\", \"split into sub-phases\", \"code review subagent\", \"parallel testing\".","spideynolove\u002Fclaude-dotfiles","spideynolove","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F19179621?v=4","01M6DPV14K2QND6S7BX6HNQYX9","Hung Nguyen","01M6DPV1MMAV9858D8R8CHM5QM",2,"MIT","https:\u002F\u002Fgithub.com\u002Fspideynolove\u002Fclaude-dotfiles\u002Fblob\u002Fmain\u002F.agents-global\u002Fskills\u002Frlm-workflow\u002Fskills\u002Frlm-subagent\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000007","代码审查","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000008","测试与质量",[27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45],"rlm subagent","subagent","rlm","dispatch","trigger","driven","plans","dispatch subagent","fallback","master","implementation","code review","use","subagents","parallel","execution","code","review","testing","2026-11-10 10:12:30",false,"---\nname: rlm-subagent\ndescription: 'Master skill for parallel subagent-driven execution with automatic fallback to single-agent sequential mode. Use when implementing plans with multiple independent sub-phases (SP1, SP2...) to dispatch parallel subagents, or when requiring code review between implementation and testing. Trigger phrases: \"parallelize\", \"dispatch subagent\", \"split into sub-phases\", \"code review subagent\", \"parallel testing\".'\n---\n\n# Subagent-Driven Execution with Fallback\n\nThis skill provides parallel subagent execution for RLM Phase 3 (Implementation), Phase 3.5 (Code Review), and Phase 4 (Testing) with automatic fallback to sequential mode when subagents are unavailable.\n\n## Trigger examples\n\n- `Parallelize Phase 3 across independent sub-phases`\n- `Dispatch an implementer subagent for each SP`\n- `Run a separate code-reviewer subagent before Phase 4`\n- `Subagents aren't available; fall back to sequential mode`\n\n## Quick Reference\n\n| Scenario | Action |\n|----------|--------|\n| Multiple independent sub-phases | Use **Parallel Mode** (subagents) |\n| Single sub-phase or subagents unavailable | Use **Sequential Mode** (fallback) |\n| Code review needed | Use **Phase 3.5** (subagent or self-review) |\n| Parallel testing | Use **Phase 4** parallel dispatch |\n\n## TODO Discipline for Subagents\n\n**The Iron Law:** NO SUBAGENT COMPLETION REPORT WITHOUT ALL TODOS CHECKED.\n\n### Subagent TODO Requirements\n\nEach subagent (implementer, code-reviewer) MUST:\n1. Include `## TODO` section in their work output\n2. Check off items as work progresses\n3. Verify ALL items checked before reporting completion\n\n### Implementer Subagent TODO Template\n\n```markdown\n## TODO\n\n- [ ] Read and understand assigned SP specification\n- [ ] Ask clarifying questions (if any)\n- [ ] Write failing test (RED phase)\n- [ ] Run test and verify failure\n- [ ] Implement minimal code (GREEN phase)\n- [ ] Run test and verify pass\n- [ ] Refactor while keeping tests green\n- [ ] Run integration tests\n- [ ] Self-review against plan\n- [ ] Document changes made\n- [ ] Report completion to controller\n```\n\n### Code Reviewer Subagent TODO Template\n\n```markdown\n## TODO\n\n- [ ] Read original plan (Phase 3)\n- [ ] Read implementation summary (Phase 3)\n- [ ] Review git diff (BASE_SHA..HEAD_SHA)\n- [ ] Verify plan alignment\n- [ ] Assess code quality\n- [ ] Check TDD compliance\n- [ ] Categorize issues (Critical\u002FImportant\u002FMinor)\n- [ ] Document positive findings\n- [ ] Render verdict\n- [ ] Report review completion\n```\n\n### Controller TODO Management\n\nThe controller MUST:\n1. Verify subagent outputs include completed TODOs\n2. NOT accept completion reports with unchecked items\n3. Request subagent to complete remaining items\n\n## Capability Detection\n\n**At start of Phase 3, detect subagent availability:**\n\n```\nIF can invoke \"agent\" or \"Task\" tool -> Use Parallel Mode\nELSE -> Use Sequential Fallback Mode\n```\n\n**Detection rule:** If the platform provides a subagent\u002Ftask primitive, use parallel mode; otherwise fallback to sequential.\n\n## Parallel Mode (Subagents Available)\n\n### Phase 3: Parallel Sub-Phase Implementation\n\n**Controller responsibilities:**\n1. Read locked Phase 2 TO-BE plan once\n2. Extract all sub-phases (SP1, SP2, SP3...)\n3. Determine dependencies (independent vs sequential)\n4. Dispatch implementer subagent per independent SP\n5. Two-stage review after each SP completion\n6. Integration testing after all approved\n\n**Dispatch pattern:**\n```typescript\n\u002F\u002F Parallel dispatch for independent SPs\nawait Promise.all([\n  Task({ description: \"Implement SP1\", prompt: implementerPrompt(SP1) }),\n  Task({ description: \"Implement SP2\", prompt: implementerPrompt(SP2) }),\n  Task({ description: \"Implement SP3\", prompt: implementerPrompt(SP3) })\n])\n```\n\n**Two-stage review:**\n1. **Spec Review:** Verify SP requirements met (plan alignment)\n2. **Code Quality Review:** Assess code quality, TDD compliance, standards\n\n### Phase 3.5: Code Review Subagent\n\n**Trigger:** After Phase 3 implementation\n**Action:** Dispatch `agents\u002Fcode-reviewer.md` subagent with:\n- BASE_SHA and HEAD_SHA of changes\n- Phase 2 TO-BE plan for alignment check\n- Severity classification (Critical\u002FImportant\u002FMinor)\n\n**Review loop:** Issues found -> implementer fixes -> re-review\n\n### Phase 4: Parallel Testing\n\n**Required before dispatching tests:**\n- Audit `03-implementation-summary.md` against `00-requirements.md` and `02-to-be-plan.md`\n- Document mismatches and remediation\u002Faddenda in Phase 4 artifact\n\n**Dispatch pattern:**\n```typescript\n\u002F\u002F Parallel test execution\nawait Promise.all([\n  Task({ description: \"Run unit tests\", prompt: testPrompt(\"unit\") }),\n  Task({ description: \"Run integration tests\", prompt: testPrompt(\"integration\") }),\n  Task({ description: \"Run E2E tests\", prompt: testPrompt(\"e2e\") })\n])\n```\n\n**Result aggregation:**\n- Collect results from all subagents\n- Summarize pass\u002Ffail counts\n- Identify any critical failures\n\n## Sequential Fallback Mode (No Subagents)\n\n**Trigger:** Subagent capability check fails\n\n**Characteristics:**\n- Execute sub-phases sequentially in main agent context\n- Extended self-review checklist per sub-phase\n- Integration testing between each sub-phase\n- Full context preservation\n\n**Fallback trigger flow:**\n```markdown\n**Subagent Check:** NOT AVAILABLE\n**Reason:** [Tool not found \u002F Platform limitation \u002F User request]\n**Action:** Using SEQUENTIAL fallback mode\n```\n\n### Sequential Execution Pattern\n\n```\nFor each SP in [SP1, SP2, SP3...]:\n  1. Execute SP implementation\n  2. Self-review against plan (extended checklist)\n  3. Document in Phase 4 artifact\n  4. Run integration tests\n  5. Proceed to next SP\n```\n\n### Sequential Review Pattern\n\n**Phase 3.5 equivalent:**\n- Main agent performs extended self-review\n- Use comprehensive checklist from `agents\u002Fcode-reviewer.md`\n- Document findings in Phase 3.5 artifact\n\n## Subagent Prompts\n\n### Implementer Subagent\n\n**File:** `agents\u002Fimplementer.md`\n\n**Key requirements:**\n- Self-contained context (full SP text provided)\n- TDD discipline enforcement\n- Question-before-work protocol\n- Self-review checklist before completion\n\n**Usage:**\n```markdown\nYou are an Implementer Agent. Implement this sub-phase:\n\n**SP Text:** [full sub-phase text]\n**BASE_SHA:** [commit SHA]\n**Context:** [relevant files]\n\nFollow the process in agents\u002Fimplementer.md\n```\n\n### Code Reviewer Subagent\n\n**File:** `agents\u002Fcode-reviewer.md`\n\n**Key requirements:**\n- Plan alignment verification\n- Code quality assessment\n- Severity classification (Critical\u002FImportant\u002FMinor)\n- Clear verdict (Approved \u002F Changes Required)\n\n**Usage:**\n```markdown\nYou are a Code Reviewer Agent. Review this implementation:\n\n**Plan:** [Phase 3 TO-BE]\n**Git Range:** [BASE_SHA..HEAD_SHA]\n**Implementation:** [Phase 4 summary]\n\nFollow the process in agents\u002Fcode-reviewer.md\n```\n\n## Artifact Documentation\n\n**Phase 4 artifact must include:**\n```markdown\n## Pre-Test Implementation Audit\n- Requirements alignment (`00-requirements.md`): [summary + evidence]\n- Plan alignment (`02-to-be-plan.md`): [summary + evidence]\n- Mismatches and remediation\u002Faddenda: [details]\n\n## Execution Mode\n- **Mode:** Parallel \u002F Sequential\n- **Subagents Used:** [names and counts]\n- **Fallback Reason:** [if applicable]\n\n## Sub-phase Results\n- SP1: [status] - [subagent name or \"main agent\"]\n- SP2: [status] - [subagent name or \"main agent\"]\n...\n\n## Review Results\n- SP1 Review: [status] - [reviewer name or \"self-review\"]\n- SP2 Review: [status] - [reviewer name or \"self-review\"]\n...\n```\n\n**Phase 3.5 artifact (if used):**\n```markdown\n## Review Scope\n- Git range: [BASE_SHA..HEAD_SHA]\n- Execution Mode: Parallel (subagent) \u002F Sequential (self-review)\n- Reviewer: [subagent name \u002F self]\n\n## Issues Found\n- Critical: [count]\n- Important: [count]\n- Minor: [count]\n\n## Verdict\n[APPROVED \u002F APPROVED WITH NOTES \u002F CHANGES REQUIRED]\n```\n\n**Phase 4 artifact must include:**\n```markdown\n## Pre-Test Implementation Audit\n- Requirements alignment (`00-requirements.md`): [summary + evidence]\n- Plan alignment (`02-to-be-plan.md`): [summary + evidence]\n- Mismatches and remediation\u002Faddenda: [details]\n\n## Execution Mode\n- **Mode:** Parallel \u002F Sequential\n- **Test Suites:**\n  - Unit: [subagent name] \u002F Main agent\n  - Integration: [subagent name] \u002F Main agent\n  - E2E: [subagent name] \u002F Main agent\n\n## Results Summary\n- Total execution time: [X] minutes\n- Estimated sequential time: [Y] minutes\n- Speedup: [Z]x\n```\n\n## Decision Tree\n\n```\nStarting Phase 4\n  |\n  v\nSubagent available? --NO--> SEQUENTIAL MODE\n  |                         - Execute SPs sequentially\n  |                         - Extended self-review\n  |                         - Document as sequential\n YES\n  |\n  v\n3+ independent SPs? --NO--> SEQUENTIAL MODE\n  |                         - Single sub-phase\n  |                         - No parallelism benefit\n YES\n  |\n  v\nPARALLEL MODE\n- Dispatch implementer per SP\n- Two-stage review\n- Integration after all done\n```\n\n## Quality Maintenance\n\n**Both modes enforce:**\n- TDD discipline\n- Artifact locking\n- Gate compliance\n- Integration testing\n\n**Parallel Mode advantages:**\n- Fresh context per sub-phase\n- Independent verification\n- Faster execution (3-5x)\n\n**Sequential Mode advantages:**\n- No subagent dependency\n- Simpler coordination\n- Full context accumulation\n\n## Common Patterns\n\n### Pattern 1: Multi-Domain Feature\n\n**Scenario:** Feature touches API, UI, and database\n\n**Phase 2 TO-BE plan:**\n```markdown\n## Sub-phases\n- SP1: API changes (backend\u002F)\n- SP2: UI changes (frontend\u002F)\n- SP3: Database migration (migrations\u002F)\n```\n\n**Phase 4 execution:**\n- SP1, SP2, SP3 dispatched in parallel (independent domains)\n- Each implementer works on isolated files\n- Integration testing after all complete\n\n### Pattern 2: Sequential Dependencies\n\n**Scenario:** SP2 depends on SP1 output\n\n**Phase 3 execution:**\n- Dispatch SP1 -> wait -> review -> approve\n- Dispatch SP2 (with SP1 context) -> wait -> review -> approve\n- Sequential, not parallel\n\n### Pattern 3: High-Risk Change\n\n**Scenario:** Critical auth system change\n\n**Phase 3:** Implement with TDD\n**Phase 3.5:** Mandatory code review\n**Phase 4:** Parallel testing (unit + integration + E2E)\n\n## Troubleshooting\n\n### Subagent fails to complete\n- Check if prompt was self-contained\n- Verify BASE_SHA was provided\n- Review logs for errors\n- Retry with clearer instructions\n\n### Integration tests fail after parallel SPs\n- Check for file conflicts between SPs\n- Verify no shared mutable state\n- Review commit history for ordering issues\n\n### Sequential mode too slow\n- Consider breaking into more granular SPs\n- Check if platform supports subagents\n- Verify tool detection is working\n\n## References\n\n- Main workflow: `SKILL.md`\n- Implementer subagent: `agents\u002Fimplementer.md`\n- Code reviewer subagent: `agents\u002Fcode-reviewer.md`\n- Artifact templates: `references\u002Fartifact-template.md`\n","# Subagent-Driven Execution with Fallback\n\nThis skill provides parallel subagent execution for RLM Phase 3 (Implementation), Phase 3.5 (Code Review), and Phase 4 (Testing) with automatic fallback to sequential mode when subagents are unavailable.\n\n## Trigger examples\n\n- `Parallelize Phase 3 across independent sub-phases`\n- `Dispatch an implementer subagent for each SP`\n- `Run a separate code-reviewer subagent before Phase 4`\n- `Subagents aren't available; fall back to sequential mode`\n\n## Quick Reference\n\n| Scenario | Action |\n|----------|--------|\n| Multiple independent sub-phases | Use **Parallel Mode** (subagents) |\n| Single sub-phase or subagents unavailable | Use **Sequential Mode** (fallback) |\n| Code review needed | Use **Phase 3.5** (subagent or self-review) |\n| Parallel testing | Use **Phase 4** parallel dispatch |\n\n## TODO Discipline for Subagents\n\n**The Iron Law:** NO SUBAGENT COMPLETION REPORT WITHOUT ALL TODOS CHECKED.\n\n### Subagent TODO Requirements\n\nEach subagent (implementer, code-reviewer) MUST:\n1. Include `## TODO` section in their work output\n2. Check off items as work progresses\n3. Verify ALL items checked before reporting completion\n\n### Implementer Subagent TODO Template\n\n```markdown\n## TODO\n\n- [ ] Read and understand assigned SP specification\n- [ ] Ask clarifying questions (if any)\n- [ ] Write failing test (RED phase)\n- [ ] Run test and verify failure\n- [ ] Implement minimal code (GREEN phase)\n- [ ] Run test and verify pass\n- [ ] Refactor while keeping tests green\n- [ ] Run integration tests\n- [ ] Self-review against plan\n- [ ] Document changes made\n- [ ] Report completion to controller\n```\n\n### Code Reviewer Subagent TODO Template\n\n```markdown\n## TODO\n\n- [ ] Read original plan (Phase 3)\n- [ ] Read implementation summary (Phase 3)\n- [ ] Review git diff (BASE_SHA..HEAD_SHA)\n- [ ] Verify plan alignment\n- [ ] Assess code quality\n- [ ] Check TDD compliance\n- [ ] Categorize issues (Critical\u002FImportant\u002FMinor)\n- [ ] Document positive findings\n- [ ] Render verdict\n- [ ] Report review completion\n```\n\n### Controller TODO Management\n\nThe controller MUST:\n1. Verify subagent outputs include completed TODOs\n2. NOT accept completion reports with unchecked items\n3. Request subagent to complete remaining items\n\n## Capability Detection\n\n**At start of Phase 3, detect subagent availability:**\n\n```\nIF can invoke \"agent\" or \"Task\" tool -> Use Parallel Mode\nELSE -> Use Sequential Fallback Mode\n```\n\n**Detection rule:** If the platform provides a subagent\u002Ftask primitive, use parallel mode; otherwise fallback to sequential.\n\n## Parallel Mode (Subagents Available)\n\n### Phase 3: Parallel Sub-Phase Implementation\n\n**Controller responsibilities:**\n1. Read locked Phase 2 TO-BE plan once\n2. Extract all sub-phases (SP1, SP2, SP3...)\n3. Determine dependencies (independent vs sequential)\n4. Dispatch implementer subagent per independent SP\n5. Two-stage review after each SP completion\n6. Integration testing after all approved\n\n**Dispatch pattern:**\n```typescript\n\u002F\u002F Parallel dispatch for independent SPs\nawait Promise.all([\n  Task({ description: \"Implement SP1\", prompt: implementerPrompt(SP1) }),\n  Task({ description: \"Implement SP2\", prompt: implementerPrompt(SP2) }),\n  Task({ description: \"Implement SP3\", prompt: implementerPrompt(SP3) })\n])\n```\n\n**Two-stage review:**\n1. **Spec Review:** Verify SP requirements met (plan alignment)\n2. **Code Quality Review:** Assess code quality, TDD compliance, standards\n\n### Phase 3.5: Code Review Subagent\n\n**Trigger:** After Phase 3 implementation\n**Action:** Dispatch `agents\u002Fcode-reviewer.md` subagent with:\n- BASE_SHA and HEAD_SHA of changes\n- Phase 2 TO-BE plan for alignment check\n- Severity classification (Critical\u002FImportant\u002FMinor)\n\n**Review loop:** Issues found -> implementer fixes -> re-review\n\n### Phase 4: Parallel Testing\n\n**Required before dispatching tests:**\n- Audit `03-implementation-summary.md` against `00-requirements.md` and `02-to-be-plan.md`\n- Document mismatches and remediation\u002Faddenda in Phase 4 artifact\n\n**Dispatch pattern:**\n```typescript\n\u002F\u002F Parallel test execution\nawait Promise.all([\n  Task({ description: \"Run unit tests\", prompt: testPrompt(\"unit\") }),\n  Task({ description: \"Run integration tests\", prompt: testPrompt(\"integration\") }),\n  Task({ description: \"Run E2E tests\", prompt: testPrompt(\"e2e\") })\n])\n```\n\n**Result aggregation:**\n- Collect results from all subagents\n- Summarize pass\u002Ffail counts\n- Identify any critical failures\n\n## Sequential Fallback Mode (No Subagents)\n\n**Trigger:** Subagent capability check fails\n\n**Characteristics:**\n- Execute sub-phases sequentially in main agent context\n- Extended self-review checklist per sub-phase\n- Integration testing between each sub-phase\n- Full context preservation\n\n**Fallback trigger flow:**\n```markdown\n**Subagent Check:** NOT AVAILABLE\n**Reason:** [Tool not found \u002F Platform limitation \u002F User request]\n**Action:** Using SEQUENTIAL fallback mode\n```\n\n### Sequential Execution Pattern\n\n```\nFor each SP in [SP1, SP2, SP3...]:\n  1. Execute SP implementation\n  2. Self-review against plan (extended checklist)\n  3. Document in Phase 4 artifact\n  4. Run integration tests\n  5. Proceed to next SP\n```\n\n### Sequential Review Pattern\n\n**Phase 3.5 equivalent:**\n- Main agent performs extended self-review\n- Use comprehensive checklist from `agents\u002Fcode-reviewer.md`\n- Document findings in Phase 3.5 artifact\n\n## Subagent Prompts\n\n### Implementer Subagent\n\n**File:** `agents\u002Fimplementer.md`\n\n**Key requirements:**\n- Self-contained context (full SP text provided)\n- TDD discipline enforcement\n- Question-before-work protocol\n- Self-review checklist before completion\n\n**Usage:**\n```markdown\nYou are an Implementer Agent. Implement this sub-phase:\n\n**SP Text:** [full sub-phase text]\n**BASE_SHA:** [commit SHA]\n**Context:** [relevant files]\n\nFollow the process in agents\u002Fimplementer.md\n```\n\n### Code Reviewer Subagent\n\n**File:** `agents\u002Fcode-reviewer.md`\n\n**Key requirements:**\n- Plan alignment verification\n- Code quality assessment\n- Severity classification (Critical\u002FImportant\u002FMinor)\n- Clear verdict (Approved \u002F Changes Required)\n\n**Usage:**\n```markdown\nYou are a Code Reviewer Agent. Review this implementation:\n\n**Plan:** [Phase 3 TO-BE]\n**Git Range:** [BASE_SHA..HEAD_SHA]\n**Implementation:** [Phase 4 summary]\n\nFollow the process in agents\u002Fcode-reviewer.md\n```\n\n## Artifact Documentation\n\n**Phase 4 artifact must include:**\n```markdown\n## Pre-Test Implementation Audit\n- Requirements alignment (`00-requirements.md`): [summary + evidence]\n- Plan alignment (`02-to-be-plan.md`): [summary + evidence]\n- Mismatches and remediation\u002Faddenda: [details]\n\n## Execution Mode\n- **Mode:** Parallel \u002F Sequential\n- **Subagents Used:** [names and counts]\n- **Fallback Reason:** [if applicable]\n\n## Sub-phase Results\n- SP1: [status] - [subagent name or \"main agent\"]\n- SP2: [status] - [subagent name or \"main agent\"]\n...\n\n## Review Results\n- SP1 Review: [status] - [reviewer name or \"self-review\"]\n- SP2 Review: [status] - [reviewer name or \"self-review\"]\n...\n```\n\n**Phase 3.5 artifact (if used):**\n```markdown\n## Review Scope\n- Git range: [BASE_SHA..HEAD_SHA]\n- Execution Mode: Parallel (subagent) \u002F Sequential (self-review)\n- Reviewer: [subagent name \u002F self]\n\n## Issues Found\n- Critical: [count]\n- Important: [count]\n- Minor: [count]\n\n## Verdict\n[APPROVED \u002F APPROVED WITH NOTES \u002F CHANGES REQUIRED]\n```\n\n**Phase 4 artifact must include:**\n```markdown\n## Pre-Test Implementation Audit\n- Requirements alignment (`00-requirements.md`): [summary + evidence]\n- Plan alignment (`02-to-be-plan.md`): [summary + evidence]\n- Mismatches and remediation\u002Faddenda: [details]\n\n## Execution Mode\n- **Mode:** Parallel \u002F Sequential\n- **Test Suites:**\n  - Unit: [subagent name] \u002F Main agent\n  - Integration: [subagent name] \u002F Main agent\n  - E2E: [subagent name] \u002F Main agent\n\n## Results Summary\n- Total execution time: [X] minutes\n- Estimated sequential time: [Y] minutes\n- Speedup: [Z]x\n```\n\n## Decision Tree\n\n```\nStarting Phase 4\n  |\n  v\nSubagent available? --NO--> SEQUENTIAL MODE\n  |                         - Execute SPs sequentially\n  |                         - Extended self-review\n  |                         - Document as sequential\n YES\n  |\n  v\n3+ independent SPs? --NO--> SEQUENTIAL MODE\n  |                         - Single sub-phase\n  |                         - No parallelism benefit\n YES\n  |\n  v\nPARALLEL MODE\n- Dispatch implementer per SP\n- Two-stage review\n- Integration after all done\n```\n\n## Quality Maintenance\n\n**Both modes enforce:**\n- TDD discipline\n- Artifact locking\n- Gate compliance\n- Integration testing\n\n**Parallel Mode advantages:**\n- Fresh context per sub-phase\n- Independent verification\n- Faster execution (3-5x)\n\n**Sequential Mode advantages:**\n- No subagent dependency\n- Simpler coordination\n- Full context accumulation\n\n## Common Patterns\n\n### Pattern 1: Multi-Domain Feature\n\n**Scenario:** Feature touches API, UI, and database\n\n**Phase 2 TO-BE plan:**\n```markdown\n## Sub-phases\n- SP1: API changes (backend\u002F)\n- SP2: UI changes (frontend\u002F)\n- SP3: Database migration (migrations\u002F)\n```\n\n**Phase 4 execution:**\n- SP1, SP2, SP3 dispatched in parallel (independent domains)\n- Each implementer works on isolated files\n- Integration testing after all complete\n\n### Pattern 2: Sequential Dependencies\n\n**Scenario:** SP2 depends on SP1 output\n\n**Phase 3 execution:**\n- Dispatch SP1 -> wait -> review -> approve\n- Dispatch SP2 (with SP1 context) -> wait -> review -> approve\n- Sequential, not parallel\n\n### Pattern 3: High-Risk Change\n\n**Scenario:** Critical auth system change\n\n**Phase 3:** Implement with TDD\n**Phase 3.5:** Mandatory code review\n**Phase 4:** Parallel testing (unit + integration + E2E)\n\n## Troubleshooting\n\n### Subagent fails to complete\n- Check if prompt was self-contained\n- Verify BASE_SHA was provided\n- Review logs for errors\n- Retry with clearer instructions\n\n### Integration tests fail after parallel SPs\n- Check for file conflicts between SPs\n- Verify no shared mutable state\n- Review commit history for ordering issues\n\n### Sequential mode too slow\n- Consider breaking into more granular SPs\n- Check if platform supports subagents\n- Verify tool detection is working\n\n## References\n\n- Main workflow: `SKILL.md`\n- Implementer subagent: `agents\u002Fimplementer.md`\n- Code reviewer subagent: `agents\u002Fcode-reviewer.md`\n- Artifact templates: `references\u002Fartifact-template.md`\n",{"name":5,"description":7},"main","be1d5a0a3cc6cb822d2451790132cda53e497525",".agents-global\u002Fskills\u002Frlm-workflow\u002Fskills\u002Frlm-subagent",3,"https:\u002F\u002Fgithub.com\u002Fspideynolove\u002Fclaude-dotfiles","https:\u002F\u002Fgithub.com\u002Fspideynolove","https:\u002F\u002Fspideynolove.github.io","Data Engineer","2026-05-28 09:03:32",{"source":8,"skillName":5,"commands":61,"prompt":71,"zipAvailable":72},[62,65,68],{"tool":63,"command":64},"npx","npx skills add spideynolove\u002Fclaude-dotfiles --skill rlm-subagent",{"tool":66,"command":67},"bunx","bunx skills add spideynolove\u002Fclaude-dotfiles --skill rlm-subagent",{"tool":69,"command":70},"pnpm","pnpm dlx skills add spideynolove\u002Fclaude-dotfiles --skill rlm-subagent","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fspideynolove\u002Fclaude-dotfiles\u002Fblob\u002Fmain\u002F.agents-global\u002Fskills\u002Frlm-workflow\u002Fskills\u002Frlm-subagent\u002FSKILL.md",true,[],[75,92,107,124,140,155],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":46},"01M6DQ1GDQVE5XE06RAQ408JGJ","xia","Borrow and adapt patterns from GitHub projects into the current codebase using a structured 6-phase protocol. Invoke with \u002Fxia \u003Cgithub-repo> [focus].","https:\u002F\u002Fgithub.com\u002Fspideynolove\u002Fclaude-dotfiles\u002Fblob\u002Fmain\u002F.opencode-global\u002Fskills\u002Fxia\u002FSKILL.md",[],[77,82,43,83,84,85,86,87,88,89,90,91],"github","protocol","projects","repo","github projects","patterns","claude-code","focus","adapt","structured",{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":46},"01M6DQ1EDYQW512EN2RFFBF4NX","x2md","Convert an X (Twitter) post URL to markdown. Use when user wants to read, summarize, archive, or analyze a tweet or X post.","https:\u002F\u002Fgithub.com\u002Fspideynolove\u002Fclaude-dotfiles\u002Fblob\u002Fmain\u002F.opencode-global\u002Fskills\u002Fx2md\u002FSKILL.md",[],[94,88,99,43,100,101,102,103,104,105,106,39],"summarize","markdown","archive","convert","analyze","twitter","read","user",{"skillNo":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":113,"publishedAt":46},"01M6DQ1BNAMJ6QGCBA17NJHZX4","sequential-thinking","Structure cognitive analysis using sequential reasoning chains. Use for codebase analysis, pattern extraction, complex problem decomposition, and building mental models before implementation.","https:\u002F\u002Fgithub.com\u002Fspideynolove\u002Fclaude-dotfiles\u002Fblob\u002Fmain\u002F.opencode-global\u002Fskills\u002Fsequential-thinking\u002FSKILL.md",[],[114,115,116,117,118,88,43,119,120,121,39,37,122,123],"sequential thinking","thinking","pattern","decomposition","problem","reasoning","extraction","analysis","models","structure",{"skillNo":125,"name":126,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":127,"categories":128,"keywords":137,"publishedAt":46},"01M6DQ17SBGVC6PQPZZTSZ49YV","rlm-workflow","https:\u002F\u002Fgithub.com\u002Fspideynolove\u002Fclaude-dotfiles\u002Fblob\u002Fmain\u002F.opencode-global\u002Fskills\u002Frlm-workflow\u002FSKILL.md",[129,132],{"categoryNo":130,"categoryName":131,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":133,"categoryName":134,"taxonomyType":135,"taxonomyTypeDesc":136},"01JBSEEDCAT000000000000037","律师与法务","PROFESSION","适用职业",[138,29,139],"rlm workflow","workflow",{"skillNo":141,"name":142,"nameZh":6,"description":143,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":144,"categories":145,"keywords":147,"publishedAt":46},"01M6DQ14RM3H655QWFXREBGC7W","repomix","Pack local or remote repos into a single context blob for whole-repository understanding. Use before tasks spanning multiple directories, architecture reviews, or onboarding to an unfamiliar codebase.","https:\u002F\u002Fgithub.com\u002Fspideynolove\u002Fclaude-dotfiles\u002Fblob\u002Fmain\u002F.opencode-global\u002Fskills\u002Frepomix\u002FSKILL.md",[146],{"categoryNo":130,"categoryName":131,"taxonomyType":21,"taxonomyTypeDesc":22},[142,148,149,150,151,43,152,153,154,39,88],"local","architecture","repos","tasks","onboarding","repository","context",{"skillNo":156,"name":157,"nameZh":6,"description":158,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":159,"categories":160,"keywords":161,"publishedAt":46},"01M6DQ12RFVYCT1YF836D5GQBA","reddit2md","Convert a Reddit post URL to markdown. Use when user wants to read, summarize, archive, or analyze a Reddit post.","https:\u002F\u002Fgithub.com\u002Fspideynolove\u002Fclaude-dotfiles\u002Fblob\u002Fmain\u002F.opencode-global\u002Fskills\u002Freddit2md\u002FSKILL.md",[],[157,100,43,162,88,39,106,163,105,103,102,99,101],"reddit post","reddit"]