[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6CY4XVCWPM09XR35VNNGN0F":3,"skill-same:01M6CY4XVCWPM09XR35VNNGN0F":82},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":38,"linkOnly":39,"markdownContent":40,"markdownBody":41,"frontMatter":42,"refName":53,"commitSha":54,"skillDirectory":55,"forksCount":13,"repositoryHtmlUrl":56,"repositoryHomepageUrl":6,"ownerHtmlUrl":57,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":58,"install":59,"resources":71},"01M6CY4XVCWPM09XR35VNNGN0F","adversarial-plan-audit",null,"Run a deep adversarial review of plans and architecture before implementation. Use when validating strategy docs, contracts, roadmaps, and competitive positioning with scored findings and prioritized recommendations.","The-Rabak\u002Fcompound-engineering-plugin","The-Rabak","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F44833742?v=4","01M6CY211HB58Y8JNWWEV9FGN1","01M6CY2165MYWKS66VN01X82EB",1,"MIT","https:\u002F\u002Fgithub.com\u002FThe-Rabak\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fportable\u002Fcompound-engineering\u002Fskills\u002Fadversarial-plan-audit\u002FSKILL.md",[],[18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37],"adversarial plan audit","audit","plan","strategy","recommendations","deep","focus","positioning","competitive positioning","implementation","kimi","use","review","openrouter","doc","gpt","run","adversarial review","architecture","codex","2026-11-10 10:12:30",false,"---\nname: adversarial-plan-audit\ndescription: Run a deep adversarial review of plans and architecture before implementation. Use when validating strategy docs, contracts, roadmaps, and competitive positioning with scored findings and prioritized recommendations.\nmodel: claude-sonnet-5\nplatforms:\n  codex:\n    model: gpt-5.5\n  copilot:\n    model: gpt-5.3-codex\n  opencode:\n    model: openrouter\u002Fmoonshotai\u002Fkimi-k2.6\nargument-hint: \"[optional: focus area, doc path, or principle]\"\ndisable-model-invocation: true\n\n---\n\n# Adversarial Plan Audit\n\nRun a rigorous pre-implementation audit for architecture and plan quality, not a superficial review. Produce a scored, evidence-cited report that pressure-tests viability, operational readiness, and differentiation.\n\n**Announce at start:** \"I'm using the adversarial-plan-audit skill to run a scored adversarial architecture and plan review.\"\n\n## When to Use\n\n- Evaluate a major architecture or implementation plan before building\n- Stress-test a memory\u002Fplugin\u002Fagent system intended for daily use\n- Decide whether a design is ready for implementation or needs redesign\n- Generate investor-grade or leadership-grade recommendation artifacts\n\n## When Not to Use\n\n- Fix isolated code bugs with clear root causes\n- Perform style-only PR review\n- Produce implementation plans from scratch without existing artifacts\n\n## Required Inputs\n\n- Primary architecture\u002Fplan\u002Fcontracts documents\n- Any benchmark\u002Fquality gates and constraints\n- Target output location for final assessment artifact\n\nIf scope is ambiguous, ask for the target decision this audit should support (go\u002Fno-go, re-scope, or rank options).\n\n## Core Workflow\n\n### Step 1: Build the Audit Matrix\n\nLoad [references\u002Faudit-rubric.md](references\u002Faudit-rubric.md) and convert scope into explicit tracks:\n\n1. Action parity\n2. Tools as primitives\n3. Context injection\n4. Shared workspace\n5. CRUD completeness\n6. UX\u002Ffeedback integration (including CLI and artifact visibility)\n7. Capability discovery and onboarding\n8. Prompt-native extensibility\n9. Adversarial risk and contradiction analysis\n10. Market positioning and differentiation\n\nApply weighting from the rubric and predefine scoring scale before collecting evidence.\n\n### Step 2: Run Parallel Adversarial Tracks\n\nLaunch track investigations in parallel with subagents.\n\nDispatch rules:\n\n1. Run up to platform concurrency limits (typically 4 concurrent). Use waves if needed.\n2. Use specialized reviewers where available for architecture, security, performance, and data integrity.\n3. Resolve local named agents before specialist dispatch:\n   - Verify the agent metadata\u002Fsource, then dispatch the resolved subagent itself.\n   - Do not include the agent file body in the dispatched prompt; pass only the audit track payload.\n4. Require each track to return:\n   - numeric score in `X\u002FY` and percentage\n   - concrete evidence with file\u002Fsection citations\n   - top gaps\n   - high-impact recommendations first\n\n### Step 3: Add Dedicated Red-Team and Market Passes\n\nAfter core principle tracks complete:\n\n1. Run red-team pass:\n   - identify likely adoption failures\n   - identify hidden coupling and contradiction risks\n   - rank by impact x likelihood\n2. Run market pass:\n   - benchmark against credible competitors\n   - classify differentiators vs commodity areas\n   - identify strategic opportunities and traps\n\n### Step 4: Synthesize, Normalize, and Prioritize\n\n1. Normalize scores into a single weighted composite.\n2. Resolve contradictions across tracks; do not average conflicting claims blindly.\n3. Produce a prioritized recommendation list using:\n   - impact on trust\u002Fadoption\n   - effort\n   - risk reduction\n   - dependency order\n4. Classify strengths separately from gaps.\n\n### Step 5: Produce a Reusable Assessment Artifact\n\nLoad [references\u002Freport-template.md](references\u002Freport-template.md). Write the full report to a persistent docs path:\n\n- Prefer `docs\u002Fassessements\u002F` if that path exists in the repository.\n- Otherwise use `docs\u002Fassessments\u002F`.\n\nName format:\n\n`YYYY-MM-DD-\u003Cproject-or-topic>-adversarial-assessment.md`\n\nThe report must include:\n\n1. Executive verdict and readiness rating\n2. Per-principle score table with status badges\n3. Adversarial risk matrix\n4. Market positioning analysis\n5. Top prioritized recommendations\n6. Clear next-step decision framing (go \u002F re-scope \u002F redesign)\n\n### Step 6: Return Tight Summary to User\n\nReturn concise summary in chat:\n\n- overall score and confidence\n- biggest 3 strengths\n- biggest 3 blockers\n- immediate next action\n\nDo not paste full report in chat when an artifact file already exists.\n\n## Output Quality Bar\n\n- Cite all critical claims to source docs or credible external references.\n- Avoid generic advice; bind recommendations to observed evidence.\n- Distinguish design defects from planned non-goals.\n- Flag uncertainty explicitly when evidence is weak.\n- Prefer smallest high-leverage interventions before large redesigns.\n\n## Failure Handling\n\n- If subagent limits block full parallelism, run in waves and continue.\n- If external research hits rate limits, proceed with available internal evidence and clearly mark confidence.\n- If required docs are missing, stop and request exact missing artifacts.\n\n## Success Criteria\n\n- [ ] Core and adversarial tracks completed\n- [ ] Numeric scored output generated per track\n- [ ] Final artifact written to repository docs path\n- [ ] Recommendations prioritized by impact\n- [ ] Summary delivered with explicit decision framing\n","# Adversarial Plan Audit\n\nRun a rigorous pre-implementation audit for architecture and plan quality, not a superficial review. Produce a scored, evidence-cited report that pressure-tests viability, operational readiness, and differentiation.\n\n**Announce at start:** \"I'm using the adversarial-plan-audit skill to run a scored adversarial architecture and plan review.\"\n\n## When to Use\n\n- Evaluate a major architecture or implementation plan before building\n- Stress-test a memory\u002Fplugin\u002Fagent system intended for daily use\n- Decide whether a design is ready for implementation or needs redesign\n- Generate investor-grade or leadership-grade recommendation artifacts\n\n## When Not to Use\n\n- Fix isolated code bugs with clear root causes\n- Perform style-only PR review\n- Produce implementation plans from scratch without existing artifacts\n\n## Required Inputs\n\n- Primary architecture\u002Fplan\u002Fcontracts documents\n- Any benchmark\u002Fquality gates and constraints\n- Target output location for final assessment artifact\n\nIf scope is ambiguous, ask for the target decision this audit should support (go\u002Fno-go, re-scope, or rank options).\n\n## Core Workflow\n\n### Step 1: Build the Audit Matrix\n\nLoad [references\u002Faudit-rubric.md](references\u002Faudit-rubric.md) and convert scope into explicit tracks:\n\n1. Action parity\n2. Tools as primitives\n3. Context injection\n4. Shared workspace\n5. CRUD completeness\n6. UX\u002Ffeedback integration (including CLI and artifact visibility)\n7. Capability discovery and onboarding\n8. Prompt-native extensibility\n9. Adversarial risk and contradiction analysis\n10. Market positioning and differentiation\n\nApply weighting from the rubric and predefine scoring scale before collecting evidence.\n\n### Step 2: Run Parallel Adversarial Tracks\n\nLaunch track investigations in parallel with subagents.\n\nDispatch rules:\n\n1. Run up to platform concurrency limits (typically 4 concurrent). Use waves if needed.\n2. Use specialized reviewers where available for architecture, security, performance, and data integrity.\n3. Resolve local named agents before specialist dispatch:\n   - Verify the agent metadata\u002Fsource, then dispatch the resolved subagent itself.\n   - Do not include the agent file body in the dispatched prompt; pass only the audit track payload.\n4. Require each track to return:\n   - numeric score in `X\u002FY` and percentage\n   - concrete evidence with file\u002Fsection citations\n   - top gaps\n   - high-impact recommendations first\n\n### Step 3: Add Dedicated Red-Team and Market Passes\n\nAfter core principle tracks complete:\n\n1. Run red-team pass:\n   - identify likely adoption failures\n   - identify hidden coupling and contradiction risks\n   - rank by impact x likelihood\n2. Run market pass:\n   - benchmark against credible competitors\n   - classify differentiators vs commodity areas\n   - identify strategic opportunities and traps\n\n### Step 4: Synthesize, Normalize, and Prioritize\n\n1. Normalize scores into a single weighted composite.\n2. Resolve contradictions across tracks; do not average conflicting claims blindly.\n3. Produce a prioritized recommendation list using:\n   - impact on trust\u002Fadoption\n   - effort\n   - risk reduction\n   - dependency order\n4. Classify strengths separately from gaps.\n\n### Step 5: Produce a Reusable Assessment Artifact\n\nLoad [references\u002Freport-template.md](references\u002Freport-template.md). Write the full report to a persistent docs path:\n\n- Prefer `docs\u002Fassessements\u002F` if that path exists in the repository.\n- Otherwise use `docs\u002Fassessments\u002F`.\n\nName format:\n\n`YYYY-MM-DD-\u003Cproject-or-topic>-adversarial-assessment.md`\n\nThe report must include:\n\n1. Executive verdict and readiness rating\n2. Per-principle score table with status badges\n3. Adversarial risk matrix\n4. Market positioning analysis\n5. Top prioritized recommendations\n6. Clear next-step decision framing (go \u002F re-scope \u002F redesign)\n\n### Step 6: Return Tight Summary to User\n\nReturn concise summary in chat:\n\n- overall score and confidence\n- biggest 3 strengths\n- biggest 3 blockers\n- immediate next action\n\nDo not paste full report in chat when an artifact file already exists.\n\n## Output Quality Bar\n\n- Cite all critical claims to source docs or credible external references.\n- Avoid generic advice; bind recommendations to observed evidence.\n- Distinguish design defects from planned non-goals.\n- Flag uncertainty explicitly when evidence is weak.\n- Prefer smallest high-leverage interventions before large redesigns.\n\n## Failure Handling\n\n- If subagent limits block full parallelism, run in waves and continue.\n- If external research hits rate limits, proceed with available internal evidence and clearly mark confidence.\n- If required docs are missing, stop and request exact missing artifacts.\n\n## Success Criteria\n\n- [ ] Core and adversarial tracks completed\n- [ ] Numeric scored output generated per track\n- [ ] Final artifact written to repository docs path\n- [ ] Recommendations prioritized by impact\n- [ ] Summary delivered with explicit decision framing\n",{"name":5,"model":43,"platforms":44,"description":7,"argument-hint":51,"disable-model-invocation":52},"claude-sonnet-5",{"codex":45,"copilot":47,"opencode":49},{"model":46},"gpt-5.5",{"model":48},"gpt-5.3-codex",{"model":50},"openrouter\u002Fmoonshotai\u002Fkimi-k2.6","[optional: focus area, doc path, or principle]",true,"main","4c70ff7e10e692dfcc6d7877a6e57dd08df94d9b","portable\u002Fcompound-engineering\u002Fskills\u002Fadversarial-plan-audit","https:\u002F\u002Fgithub.com\u002FThe-Rabak\u002Fcompound-engineering-plugin","https:\u002F\u002Fgithub.com\u002FThe-Rabak","2026-07-28 15:25:35",{"source":8,"skillName":5,"commands":60,"prompt":70,"zipAvailable":52},[61,64,67],{"tool":62,"command":63},"npx","npx skills add The-Rabak\u002Fcompound-engineering-plugin --skill adversarial-plan-audit",{"tool":65,"command":66},"bunx","bunx skills add The-Rabak\u002Fcompound-engineering-plugin --skill adversarial-plan-audit",{"tool":68,"command":69},"pnpm","pnpm dlx skills add The-Rabak\u002Fcompound-engineering-plugin --skill adversarial-plan-audit","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FThe-Rabak\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fportable\u002Fcompound-engineering\u002Fskills\u002Fadversarial-plan-audit\u002FSKILL.md",[72,78],{"path":73,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":76,"sourceHtmlUrl":77},"references\u002Faudit-rubric.md","REFERENCE","参考文档",2560,"https:\u002F\u002Fgithub.com\u002FThe-Rabak\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fportable\u002Fcompound-engineering\u002Fskills\u002Fadversarial-plan-audit\u002Freferences\u002Faudit-rubric.md",{"path":79,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":80,"sourceHtmlUrl":81},"references\u002Freport-template.md",1831,"https:\u002F\u002Fgithub.com\u002FThe-Rabak\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fportable\u002Fcompound-engineering\u002Fskills\u002Fadversarial-plan-audit\u002Freferences\u002Freport-template.md",[83,99,108,118,126,133],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":94,"publishedAt":38},"01M6CY858Z5RCC30TKAVBDJ6SZ","skill-one","Shared reusable skill.","https:\u002F\u002Fgithub.com\u002FThe-Rabak\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsample-portable-plugin\u002Fskills\u002Fskill-one\u002FSKILL.md",[89],{"categoryNo":90,"categoryName":91,"taxonomyType":92,"taxonomyTypeDesc":93},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[95,96,97,98,33],"skill one","shared","reusable","haiku",{"skillNo":100,"name":85,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":105,"publishedAt":38},"01M6CY843KQG3FQF6259420CCD","Sample skill","https:\u002F\u002Fgithub.com\u002FThe-Rabak\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsample-plugin\u002Fskills\u002Fskill-one\u002FSKILL.md",[104],{"categoryNo":90,"categoryName":91,"taxonomyType":92,"taxonomyTypeDesc":93},[95,106,107],"sample","sample skill",{"skillNo":109,"name":110,"nameZh":6,"description":111,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":112,"categories":113,"keywords":115,"publishedAt":38},"01M6CY82607GGPWR5H25SY74H2","disabled-skill","A skill with model invocation disabled","https:\u002F\u002Fgithub.com\u002FThe-Rabak\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsample-plugin\u002Fskills\u002Fdisabled-skill\u002FSKILL.md",[114],{"categoryNo":90,"categoryName":91,"taxonomyType":92,"taxonomyTypeDesc":93},[116,117],"disabled skill","model",{"skillNo":119,"name":120,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":121,"categories":122,"keywords":123,"publishedAt":38},"01M6CY80RWMK81T97A2Q5KAFAF","default-skill","https:\u002F\u002Fgithub.com\u002FThe-Rabak\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fcustom-paths\u002Fskills\u002Fdefault-skill\u002FSKILL.md",[],[124,125],"default skill","default",{"skillNo":127,"name":128,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":129,"categories":130,"keywords":131,"publishedAt":38},"01M6CY7ZK2PMV75ES90W6YYD4S","custom-skill","https:\u002F\u002Fgithub.com\u002FThe-Rabak\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fcustom-paths\u002Fcustom-skills\u002Fcustom-skill\u002FSKILL.md",[],[132],"custom skill",{"skillNo":134,"name":135,"nameZh":6,"description":136,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":137,"categories":138,"keywords":140,"publishedAt":38},"01M6CY7V6C7HFCNJEVNFT1X00T","workflow-next-step","This skill should be used at the end of core compound-engineering workflows to inspect completed artifacts, show the workflow progress checklist, and give the exact next-session command with inputs.","https:\u002F\u002Fgithub.com\u002FThe-Rabak\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fportable\u002Fcompound-engineering\u002Fskills\u002Fworkflow-next-step\u002FSKILL.md",[139],{"categoryNo":90,"categoryName":91,"taxonomyType":92,"taxonomyTypeDesc":93},[141,142,143,144,145,146,147,148,149,150,151,152],"workflow","next","workflow next step","compound","show","checklist","session","command","engineering","artifacts","core","workflows"]