{
  "schemaVersion": "1.0",
  "item": {
    "slug": "adversarial-coach",
    "name": "adversarial-coach",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/killerapp/adversarial-coach",
    "canonicalUrl": "https://clawhub.ai/killerapp/adversarial-coach",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/adversarial-coach",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=adversarial-coach",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md"
    ],
    "primaryDoc": "SKILL.md",
    "quickSetup": [
      "Download the package from Yavira.",
      "Extract the archive and review SKILL.md first.",
      "Import or place the package into your OpenClaw setup."
    ],
    "agentAssist": {
      "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.",
      "steps": [
        "Download the package from Yavira.",
        "Extract it into a folder your agent can access.",
        "Paste one of the prompts below and point your agent at the extracted folder."
      ],
      "prompts": [
        {
          "label": "New install",
          "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete."
        },
        {
          "label": "Upgrade existing",
          "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run."
        }
      ]
    },
    "sourceHealth": {
      "source": "tencent",
      "slug": "adversarial-coach",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-11T15:10:52.585Z",
      "expiresAt": "2026-05-18T15:10:52.585Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=adversarial-coach",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=adversarial-coach",
        "contentDisposition": "attachment; filename=\"adversarial-coach-0.9.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "adversarial-coach"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/adversarial-coach"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    },
    "downloadPageUrl": "https://openagent3.xyz/downloads/adversarial-coach",
    "agentPageUrl": "https://openagent3.xyz/skills/adversarial-coach/agent",
    "manifestUrl": "https://openagent3.xyz/skills/adversarial-coach/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/adversarial-coach/agent.md"
  },
  "agentAssist": {
    "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.",
    "steps": [
      "Download the package from Yavira.",
      "Extract it into a folder your agent can access.",
      "Paste one of the prompts below and point your agent at the extracted folder."
    ],
    "prompts": [
      {
        "label": "New install",
        "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete."
      },
      {
        "label": "Upgrade existing",
        "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run."
      }
    ]
  },
  "documentation": {
    "source": "clawhub",
    "primaryDoc": "SKILL.md",
    "sections": [
      {
        "title": "Usage",
        "body": "/coach [requirements-file]\n\n/coach - Infer requirements from context\n/coach requirements.md - Validate against specific file"
      },
      {
        "title": "Coach-Player Loop",
        "body": "You orchestrate this dialectical loop between implementing agent (player) and reviewer (coach):\n\nYou (player) implement features\n/coach invokes adversarial review with independent evaluation of compliance to requirements\nCoach returns: IMPLEMENTATION_APPROVED or specific fixes\nAddress feedback, loop until approved"
      },
      {
        "title": "Step 1: Identify Requirements",
        "body": "Check (in order):\n\nSpecified requirements file or issue/ticket mentioned\nrequirements.md, REQUIREMENTS.md, SPEC.md, TODO.md\nConversation context; ask user if nothing found"
      },
      {
        "title": "Step 2: Adversarial Review",
        "body": "Review with fresh objectivity - discard prior knowledge, don't rationalize shortcuts.\n\nCheck CategoryItemsRequirementsEach item: implemented or missing with specific gapCompilationCompiles? Tests pass? Runs?Common GapsAuth on endpoints, token refresh endpoint, HTTPS, bcrypt for passwords, error handling, input validationFunctionalTest actual flows (not just compilation), verify edge cases workTest CoverageAuth error cases (401/403), token expiry, invalid inputs, rate limits"
      },
      {
        "title": "Step 3: Return Verdict",
        "body": "If approved (>95% complete):\n\nIMPLEMENTATION_APPROVED\n\n- [Requirement 1]: Verified\n- [Requirement 2]: Verified\n- Compilation: Success\n- Tests: All passing\n\nIf fixes needed:\n\nREQUIREMENTS COMPLIANCE:\n- [Requirement]: Implemented\n- [Requirement]: Missing - [specific gap]\n\nIMMEDIATE ACTIONS NEEDED:\n1. [Specific fix with file/line if known]\n2. [Specific fix]"
      },
      {
        "title": "Key Principles",
        "body": "Rigorous but fair:\n\nCatch real gaps (security, logic, data flow), not style preferences\nFunctionality over aesthetics; always flag security issues (auth, crypto, validation)\n\nConcise:\n\nBullets, not essays; specific issues, not vague concerns\nNo file contents or verbose analysis in output\n\nFresh context is your superpower:\n\nReview as if you've never seen this code; validate against requirements, not intentions"
      },
      {
        "title": "Magic Signal",
        "body": "IMPLEMENTATION_APPROVED = termination signal\n\nOnly use when: all requirements met, code compiles/runs, tests pass, no significant gaps.\n\nIf in doubt, don't approve."
      },
      {
        "title": "Minimal Example",
        "body": "User: /coach SPEC.md\n\nClaude:\nREQUIREMENTS COMPLIANCE:\n- User registration endpoint (bcrypt hashing)\n- Login with JWT tokens\n- Logout endpoint - not implemented\n- Token refresh - missing /auth/refresh endpoint\n\nIMMEDIATE ACTIONS NEEDED:\n1. Add POST /auth/logout endpoint\n2. Add POST /auth/refresh for token renewal\n3. Add tests for 401 on expired tokens\n\n---\n\nUser: [fixes] /coach SPEC.md\n\nClaude:\nIMPLEMENTATION_APPROVED\n- All auth endpoints verified (register, login, logout, refresh)\n- 18 tests passing including auth error cases"
      },
      {
        "title": "Research",
        "body": "Paper: Adversarial Cooperation in Code Synthesis\nImplementation: g3\nKey insight: Discard implementing agent's self-report; perform independent evaluation against requirements."
      }
    ],
    "body": "/coach - Adversarial Implementation Review\nUsage\n/coach [requirements-file]\n\n/coach - Infer requirements from context\n/coach requirements.md - Validate against specific file\nCoach-Player Loop\n\nYou orchestrate this dialectical loop between implementing agent (player) and reviewer (coach):\n\nYou (player) implement features\n/coach invokes adversarial review with independent evaluation of compliance to requirements\nCoach returns: IMPLEMENTATION_APPROVED or specific fixes\nAddress feedback, loop until approved\nReview Process\nStep 1: Identify Requirements\n\nCheck (in order):\n\nSpecified requirements file or issue/ticket mentioned\nrequirements.md, REQUIREMENTS.md, SPEC.md, TODO.md\nConversation context; ask user if nothing found\nStep 2: Adversarial Review\n\nReview with fresh objectivity - discard prior knowledge, don't rationalize shortcuts.\n\nCheck Category\tItems\nRequirements\tEach item: implemented or missing with specific gap\nCompilation\tCompiles? Tests pass? Runs?\nCommon Gaps\tAuth on endpoints, token refresh endpoint, HTTPS, bcrypt for passwords, error handling, input validation\nFunctional\tTest actual flows (not just compilation), verify edge cases work\nTest Coverage\tAuth error cases (401/403), token expiry, invalid inputs, rate limits\nStep 3: Return Verdict\n\nIf approved (>95% complete):\n\nIMPLEMENTATION_APPROVED\n\n- [Requirement 1]: Verified\n- [Requirement 2]: Verified\n- Compilation: Success\n- Tests: All passing\n\n\nIf fixes needed:\n\nREQUIREMENTS COMPLIANCE:\n- [Requirement]: Implemented\n- [Requirement]: Missing - [specific gap]\n\nIMMEDIATE ACTIONS NEEDED:\n1. [Specific fix with file/line if known]\n2. [Specific fix]\n\nKey Principles\n\nRigorous but fair:\n\nCatch real gaps (security, logic, data flow), not style preferences\nFunctionality over aesthetics; always flag security issues (auth, crypto, validation)\n\nConcise:\n\nBullets, not essays; specific issues, not vague concerns\nNo file contents or verbose analysis in output\n\nFresh context is your superpower:\n\nReview as if you've never seen this code; validate against requirements, not intentions\nMagic Signal\n\nIMPLEMENTATION_APPROVED = termination signal\n\nOnly use when: all requirements met, code compiles/runs, tests pass, no significant gaps.\n\nIf in doubt, don't approve.\n\nMinimal Example\nUser: /coach SPEC.md\n\nClaude:\nREQUIREMENTS COMPLIANCE:\n- User registration endpoint (bcrypt hashing)\n- Login with JWT tokens\n- Logout endpoint - not implemented\n- Token refresh - missing /auth/refresh endpoint\n\nIMMEDIATE ACTIONS NEEDED:\n1. Add POST /auth/logout endpoint\n2. Add POST /auth/refresh for token renewal\n3. Add tests for 401 on expired tokens\n\n---\n\nUser: [fixes] /coach SPEC.md\n\nClaude:\nIMPLEMENTATION_APPROVED\n- All auth endpoints verified (register, login, logout, refresh)\n- 18 tests passing including auth error cases\n\nResearch\nPaper: Adversarial Cooperation in Code Synthesis\nImplementation: g3\nKey insight: Discard implementing agent's self-report; perform independent evaluation against requirements."
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/killerapp/adversarial-coach",
    "publisherUrl": "https://clawhub.ai/killerapp/adversarial-coach",
    "owner": "killerapp",
    "version": "0.9.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/adversarial-coach",
    "downloadUrl": "https://openagent3.xyz/downloads/adversarial-coach",
    "agentUrl": "https://openagent3.xyz/skills/adversarial-coach/agent",
    "manifestUrl": "https://openagent3.xyz/skills/adversarial-coach/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/adversarial-coach/agent.md"
  }
}