{
  "schemaVersion": "1.0",
  "item": {
    "slug": "frankenstein",
    "name": "Frankenstein",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/RubenAQuispe/frankenstein",
    "canonicalUrl": "https://clawhub.ai/RubenAQuispe/frankenstein",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/frankenstein",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=frankenstein",
    "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",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-30T16:55:25.780Z",
      "expiresAt": "2026-05-07T16:55:25.780Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network",
        "contentDisposition": "attachment; filename=\"network-1.0.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null
      },
      "scope": "source",
      "summary": "Source download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this source.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/frankenstein"
    },
    "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/frankenstein",
    "agentPageUrl": "https://openagent3.xyz/skills/frankenstein/agent",
    "manifestUrl": "https://openagent3.xyz/skills/frankenstein/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/frankenstein/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": "Model Requirements",
        "body": "Default: Opus (or best available thinking model)\n\nFrankenstein requires deep reasoning to:\n\nCompare multiple skill approaches\nIdentify subtle methodology differences\nSynthesize the best parts creatively\nCatch security/quality issues others miss\n\nOnly use a smaller model if user explicitly requests it for cost reasons. The synthesis quality depends heavily on reasoning depth.\n\nCreate monster skills by combining the best parts of existing ones."
      },
      {
        "title": "Quick Start",
        "body": "Frankenstein me an SEO audit skill"
      },
      {
        "title": "Step 1: Search All Sources",
        "body": "Search EVERY AI skills repository for matching skills:\n\n1. ClawHub (primary)\n\nclawhub search \"[topic]\" --registry \"https://clawhub.ai\"\n\n2. GitHub\n\nSearch: \"[topic] AI skill\" OR \"[topic] claude skill\" OR \"[topic] agent skill\"\nLook for: SKILL.md, CLAUDE.md, or similar agent instruction files\n\n3. skills.sh\n\nhttps://skills.sh/search?q=[topic]\n\n4. skillsmp.com (Skills Marketplace)\n\nhttps://skillsmp.com/search/[topic]\n\n5. Other sources to check:\n\nAnthropic's skill examples\nOpenAI GPT configurations (convert to skill format)\nLangChain agent templates\nAutoGPT/AgentGPT skill repos\n\nGather all candidates before filtering. More sources = better Frankenstein."
      },
      {
        "title": "Step 2: Security Scan",
        "body": "Run each skill through skill-auditor. Skip any with HIGH risk scores.\n\nFor each skill found:\n\nInstall to temp directory\nRun skill-auditor scan\nScore >= 7 = SAFE (proceed)\nScore < 7 = RISKY (skip with warning)"
      },
      {
        "title": "Step 3: Safe Analysis",
        "body": "Analyze safe skills in sandwrap read-only mode.\n\nFor each safe skill, extract:\n\nCore features (what it does)\nMethodology (how it approaches the problem)\nScripts/tools (reusable code)\nUnique strengths (what makes it special)\nWeaknesses (what's missing)"
      },
      {
        "title": "Step 4: Compare",
        "body": "Build comparison matrix:\n\nFeatureskill-Askill-Bskill-CWINNERFeature 1YesNoYesA, CFeature 2BasicAdvancedNoneBFeature 3NoNoYesC"
      },
      {
        "title": "Step 5: Synthesize",
        "body": "Take the winning approach for each feature:\n\nFeature 1 methodology from skill-A\nFeature 2 implementation from skill-B\nFeature 3 approach from skill-C"
      },
      {
        "title": "Step 6: Build Initial Draft",
        "body": "Use skill-creator to assemble the Frankenstein skill:\n\nCombine winning features\nResolve conflicts (if two approaches clash)\nWrite unified SKILL.md\nInclude scripts from winners\nDocument sources"
      },
      {
        "title": "Step 7: Vetting Loop (CRITICAL)",
        "body": "Run plan → test → improve loop until 3 stable passes:\n\nPass 1:\n  1. Read draft\n  2. Try to break it (find holes, contradictions, gaps)\n  3. Document issues\n  4. Fix them\n  \nPass 2:\n  1. Read improved version\n  2. Actively try to find MORE issues\n  3. Fix any found\n  \nPass 3+:\n  Continue until you genuinely try to improve\n  but can't find significant issues\n\nWhat to look for each pass:\n\nMissing features that sources had\nContradictions between combined approaches\nVague instructions that aren't actionable\nToken waste (verbose where concise works)\nSecurity gaps\nBroken references to files/scripts\n\nDocument in VETTING-LOG.md:\n\nEach pass number\nIssues found\nFixes applied\nWhy considered stable\n\nOnly proceed when:\n\n3 consecutive passes with no major issues\nMinor issues documented as known limitations"
      },
      {
        "title": "Step 8: Human Review",
        "body": "Present the vetted skill for approval:\n\nShow what came from where\nHighlight conflicts resolved\nShow vetting summary\nAsk for final OK before saving"
      },
      {
        "title": "Output",
        "body": "Creates a new skill with:\n\nBest features from all analyzed skills\nClear attribution (credits source skills)\nSecurity-scanned components only\nUnified documentation"
      },
      {
        "title": "Example Session",
        "body": "User: \"Frankenstein me an SEO audit skill\"\n\nFrankenstein:\n\nSearching ClawHub for \"SEO audit\"...\nFound 5 skills:\n  1. seo-audit (coreyhaines31)\n  2. audit-website (squirrelscan)\n  3. seo-optimizer (local)\n  4. technical-seo (unknown)\n  5. seo-checker (unknown)\n\nSecurity scanning...\n  ✓ seo-audit: SAFE (8/10)\n  ✓ audit-website: SAFE (7/10)\n  ✓ seo-optimizer: SAFE (9/10)\n  ✗ technical-seo: RISKY (4/10) - skipping\n  ✗ seo-checker: RISKY (3/10) - skipping\n\nAnalyzing 3 safe skills...\n\nComparison:\n| Feature              | seo-audit | audit-website | seo-optimizer |\n|----------------------|-----------|---------------|---------------|\n| Methodology depth    | ★★★★★     | ★★★           | ★★★★          |\n| Automated rules      | None      | 230+          | 50+           |\n| Auto-fix capability  | No        | No            | Yes           |\n| CLI tool             | No        | squirrel      | No            |\n| E-E-A-T coverage     | Yes       | Yes           | Partial       |\n\nRecommended combination:\n- Methodology: seo-audit (best framework)\n- Rules engine: audit-website (230+ rules via squirrel CLI)\n- Auto-fix: seo-optimizer (automated fixes)\n- E-E-A-T: seo-audit (comprehensive)\n\nBuild this Frankenstein? [Yes/No]"
      },
      {
        "title": "Dependencies",
        "body": "This skill uses:\n\nclawhub CLI (search/install)\nskill-auditor (security scanning)\nsandwrap (safe analysis)\nskill-creator (building)"
      },
      {
        "title": "Spawning Sub-Agents",
        "body": "When spawning analysis sub-agents, always use Opus (or best thinking model) unless user explicitly requests otherwise:\n\nsessions_spawn(\n  task: \"FRANKENSTEIN ANALYSIS: [topic]...\",\n  model: \"opus\"\n)\n\nCheaper models miss nuances between skills and produce shallow combinations."
      },
      {
        "title": "Limitations",
        "body": "Only combines publicly available skills\nSkips skills that fail security scan\nCannot resolve deep architectural conflicts\nHuman judgment needed for final synthesis\nQuality depends on available skills"
      },
      {
        "title": "Credits",
        "body": "When a Frankenstein skill is built, it includes attribution:\n\n## Sources\nBuilt from best parts of:\n- seo-audit by coreyhaines31 (methodology)\n- audit-website by squirrelscan (rules engine)\n- seo-optimizer (auto-fix)"
      }
    ],
    "body": "Frankenstein\nModel Requirements\n\nDefault: Opus (or best available thinking model)\n\nFrankenstein requires deep reasoning to:\n\nCompare multiple skill approaches\nIdentify subtle methodology differences\nSynthesize the best parts creatively\nCatch security/quality issues others miss\n\nOnly use a smaller model if user explicitly requests it for cost reasons. The synthesis quality depends heavily on reasoning depth.\n\nCreate monster skills by combining the best parts of existing ones.\n\nQuick Start\nFrankenstein me an SEO audit skill\n\nHow It Works\nStep 1: Search All Sources\n\nSearch EVERY AI skills repository for matching skills:\n\n1. ClawHub (primary)\n\nclawhub search \"[topic]\" --registry \"https://clawhub.ai\"\n\n\n2. GitHub\n\nSearch: \"[topic] AI skill\" OR \"[topic] claude skill\" OR \"[topic] agent skill\"\nLook for: SKILL.md, CLAUDE.md, or similar agent instruction files\n\n\n3. skills.sh\n\nhttps://skills.sh/search?q=[topic]\n\n\n4. skillsmp.com (Skills Marketplace)\n\nhttps://skillsmp.com/search/[topic]\n\n\n5. Other sources to check:\n\nAnthropic's skill examples\nOpenAI GPT configurations (convert to skill format)\nLangChain agent templates\nAutoGPT/AgentGPT skill repos\n\nGather all candidates before filtering. More sources = better Frankenstein.\n\nStep 2: Security Scan\n\nRun each skill through skill-auditor. Skip any with HIGH risk scores.\n\nFor each skill found:\n\nInstall to temp directory\nRun skill-auditor scan\nScore >= 7 = SAFE (proceed)\nScore < 7 = RISKY (skip with warning)\nStep 3: Safe Analysis\n\nAnalyze safe skills in sandwrap read-only mode.\n\nFor each safe skill, extract:\n\nCore features (what it does)\nMethodology (how it approaches the problem)\nScripts/tools (reusable code)\nUnique strengths (what makes it special)\nWeaknesses (what's missing)\nStep 4: Compare\n\nBuild comparison matrix:\n\nFeature\tskill-A\tskill-B\tskill-C\tWINNER\nFeature 1\tYes\tNo\tYes\tA, C\nFeature 2\tBasic\tAdvanced\tNone\tB\nFeature 3\tNo\tNo\tYes\tC\nStep 5: Synthesize\n\nTake the winning approach for each feature:\n\nFeature 1 methodology from skill-A\nFeature 2 implementation from skill-B\nFeature 3 approach from skill-C\nStep 6: Build Initial Draft\n\nUse skill-creator to assemble the Frankenstein skill:\n\nCombine winning features\nResolve conflicts (if two approaches clash)\nWrite unified SKILL.md\nInclude scripts from winners\nDocument sources\nStep 7: Vetting Loop (CRITICAL)\n\nRun plan → test → improve loop until 3 stable passes:\n\nPass 1:\n  1. Read draft\n  2. Try to break it (find holes, contradictions, gaps)\n  3. Document issues\n  4. Fix them\n  \nPass 2:\n  1. Read improved version\n  2. Actively try to find MORE issues\n  3. Fix any found\n  \nPass 3+:\n  Continue until you genuinely try to improve\n  but can't find significant issues\n\n\nWhat to look for each pass:\n\nMissing features that sources had\nContradictions between combined approaches\nVague instructions that aren't actionable\nToken waste (verbose where concise works)\nSecurity gaps\nBroken references to files/scripts\n\nDocument in VETTING-LOG.md:\n\nEach pass number\nIssues found\nFixes applied\nWhy considered stable\n\nOnly proceed when:\n\n3 consecutive passes with no major issues\nMinor issues documented as known limitations\nStep 8: Human Review\n\nPresent the vetted skill for approval:\n\nShow what came from where\nHighlight conflicts resolved\nShow vetting summary\nAsk for final OK before saving\nOutput\n\nCreates a new skill with:\n\nBest features from all analyzed skills\nClear attribution (credits source skills)\nSecurity-scanned components only\nUnified documentation\nExample Session\n\nUser: \"Frankenstein me an SEO audit skill\"\n\nFrankenstein:\n\nSearching ClawHub for \"SEO audit\"...\nFound 5 skills:\n  1. seo-audit (coreyhaines31)\n  2. audit-website (squirrelscan)\n  3. seo-optimizer (local)\n  4. technical-seo (unknown)\n  5. seo-checker (unknown)\n\nSecurity scanning...\n  ✓ seo-audit: SAFE (8/10)\n  ✓ audit-website: SAFE (7/10)\n  ✓ seo-optimizer: SAFE (9/10)\n  ✗ technical-seo: RISKY (4/10) - skipping\n  ✗ seo-checker: RISKY (3/10) - skipping\n\nAnalyzing 3 safe skills...\n\nComparison:\n| Feature              | seo-audit | audit-website | seo-optimizer |\n|----------------------|-----------|---------------|---------------|\n| Methodology depth    | ★★★★★     | ★★★           | ★★★★          |\n| Automated rules      | None      | 230+          | 50+           |\n| Auto-fix capability  | No        | No            | Yes           |\n| CLI tool             | No        | squirrel      | No            |\n| E-E-A-T coverage     | Yes       | Yes           | Partial       |\n\nRecommended combination:\n- Methodology: seo-audit (best framework)\n- Rules engine: audit-website (230+ rules via squirrel CLI)\n- Auto-fix: seo-optimizer (automated fixes)\n- E-E-A-T: seo-audit (comprehensive)\n\nBuild this Frankenstein? [Yes/No]\n\nDependencies\n\nThis skill uses:\n\nclawhub CLI (search/install)\nskill-auditor (security scanning)\nsandwrap (safe analysis)\nskill-creator (building)\nSpawning Sub-Agents\n\nWhen spawning analysis sub-agents, always use Opus (or best thinking model) unless user explicitly requests otherwise:\n\nsessions_spawn(\n  task: \"FRANKENSTEIN ANALYSIS: [topic]...\",\n  model: \"opus\"\n)\n\n\nCheaper models miss nuances between skills and produce shallow combinations.\n\nLimitations\nOnly combines publicly available skills\nSkips skills that fail security scan\nCannot resolve deep architectural conflicts\nHuman judgment needed for final synthesis\nQuality depends on available skills\nCredits\n\nWhen a Frankenstein skill is built, it includes attribution:\n\n## Sources\nBuilt from best parts of:\n- seo-audit by coreyhaines31 (methodology)\n- audit-website by squirrelscan (rules engine)\n- seo-optimizer (auto-fix)"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/RubenAQuispe/frankenstein",
    "publisherUrl": "https://clawhub.ai/RubenAQuispe/frankenstein",
    "owner": "RubenAQuispe",
    "version": "1.2.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/frankenstein",
    "downloadUrl": "https://openagent3.xyz/downloads/frankenstein",
    "agentUrl": "https://openagent3.xyz/skills/frankenstein/agent",
    "manifestUrl": "https://openagent3.xyz/skills/frankenstein/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/frankenstein/agent.md"
  }
}