{
  "schemaVersion": "1.0",
  "item": {
    "slug": "guard-scanner",
    "name": "guard-scanner",
    "source": "tencent",
    "type": "skill",
    "category": "安全合规",
    "sourceUrl": "https://clawhub.ai/koatora20/guard-scanner",
    "canonicalUrl": "https://clawhub.ai/koatora20/guard-scanner",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/guard-scanner",
    "sourceDownloadUrl": "https://clawhub.ai/api/v1/download?slug=guard-scanner",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "CHANGELOG.md",
      "CODE_OF_CONDUCT.md",
      "CONTRIBUTING.md",
      "GOVERNANCE.md",
      "MAINTAINERS.md",
      "README.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. Then review README.md for any prerequisites, environment setup, or post-install checks. 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. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run."
        }
      ]
    },
    "sourceHealth": {
      "source": "tencent",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-23T16:43:11.935Z",
      "expiresAt": "2026-04-30T16:43:11.935Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
        "contentDisposition": "attachment; filename=\"4claw-imageboard-1.0.1.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/guard-scanner"
    },
    "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/guard-scanner",
    "agentPageUrl": "https://openagent3.xyz/skills/guard-scanner/agent",
    "manifestUrl": "https://openagent3.xyz/skills/guard-scanner/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/guard-scanner/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. Then review README.md for any prerequisites, environment setup, or post-install checks. 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. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run."
      }
    ]
  },
  "documentation": {
    "source": "clawhub",
    "primaryDoc": "SKILL.md",
    "sections": [
      {
        "title": "guard-scanner",
        "body": "Scan AI agent skills for 35 categories of threats. v16 adds a 5-layer analysis pipeline, OWASP ASI projection mode, richer finding metadata, and Rust runtime evidence integration on top of the existing prompt injection, identity hijacking, memory poisoning, MCP poisoning, and supply chain coverage."
      },
      {
        "title": "Quick Start",
        "body": "# Scan a skill directory\nnpx -y @guava-parity/guard-scanner ./my-skills/ --verbose\n\n# Scan with identity protection\nnpx -y @guava-parity/guard-scanner ./skills/ --soul-lock --strict\n\n# Filter to OWASP ASI mapped findings only\nnpx -y @guava-parity/guard-scanner ./skills/ --compliance owasp-asi --format json\n\n# Installed CLI\nguard-scanner ./skills/ --strict\n\n# npm exec compatibility\nnpm exec --yes --package=@guava-parity/guard-scanner -- guard-scanner ./skills/ --strict"
      },
      {
        "title": "Scan",
        "body": "guard-scanner <dir>         # Scan directory\nguard-scanner <dir> -v      # Verbose output\nguard-scanner <dir> --json  # JSON report file\nguard-scanner <dir> --sarif # SARIF for CI/CD\nguard-scanner <dir> --html  # HTML report\nguard-scanner <dir> --compliance owasp-asi --format json"
      },
      {
        "title": "Asset Audit",
        "body": "Audit public registries for credential exposure.\n\nguard-scanner audit npm <username>\nguard-scanner audit github <username>\nguard-scanner audit clawhub <query>\nguard-scanner audit all <username> --verbose"
      },
      {
        "title": "MCP Server",
        "body": "Start as MCP server for IDE integration.\n\nguard-scanner serve\n\nEditor config (Cursor, Windsurf, Claude Code, OpenClaw):\n\n{\n  \"mcpServers\": {\n    \"guard-scanner\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@guava-parity/guard-scanner\", \"serve\"]\n    }\n  }\n}\n\nMCP tools: scan_skill, scan_text, check_tool_call, audit_assets, get_stats, and the async experimental task helpers."
      },
      {
        "title": "Quality Contract",
        "body": "Public quality contract:\n\nBenchmark corpus version: 2026-03-13.quality-v1\nPrecision target: >= 0.90\nRecall target: >= 0.90\nFPR/FNR budgets: <= 0.10\nExplainability completeness: 1.0\nRuntime policy latency budget: 5ms\n\nEvidence surfaces:\n\ndocs/spec/capabilities.json\ndocs/data/corpus-metrics.json\ndocs/data/benchmark-ledger.json\ndocs/data/fp-ledger.json"
      },
      {
        "title": "Watch Mode",
        "body": "Monitor skill directories in real-time during development.\n\nguard-scanner watch ./skills/ --strict --soul-lock"
      },
      {
        "title": "VirusTotal Integration",
        "body": "Combine semantic detection with VirusTotal's 70+ antivirus engines. Optional — guard-scanner works fully without it.\n\nexport VT_API_KEY=your-key\nguard-scanner scan ./skills/ --vt-scan"
      },
      {
        "title": "Runtime Guard",
        "body": "The validated OpenClaw surface is the compiled runtime plugin entry (dist/openclaw-plugin.mjs) discovered through package.json > openclaw.extensions and mounted on before_tool_call for OpenClaw v2026.3.13, with regression coverage kept on v2026.3.8.\n\nThe before_tool_call hook provides 27 runtime checks across 5 defense layers, while v16 scan output adds a second 5-layer analysis view:\n\nLayerFocus1. Threat DetectionReverse shell, curl|bash, SSRF2. Trust DefenseSOUL.md tampering, memory injection3. Safety JudgePrompt injection in tool arguments4. BehavioralNo-research execution detection5. Trust ExploitationAuthority claims, creator bypass\n\nModes: monitor (log only), enforce (block CRITICAL, default), strict (block HIGH+)."
      },
      {
        "title": "v16 Output Surface",
        "body": "Finding fields: layer, layer_name, owasp_asi, protocol_surface\nCompliance mode: --compliance owasp-asi\nMCP summaries: scan_skill, scan_text, and get_stats now surface layer and ASI context\nRuntime evidence: Rust memory_integrity and soul_hard_gate modules are represented in the TypeScript pipeline"
      },
      {
        "title": "Key Flags",
        "body": "FlagEffect--verbose / -vDetailed findings with line numbers--strictLower detection thresholds--soul-lockEnable identity protection patterns--json / --sarif / --htmlOutput format--fail-on-findingsExit 1 on findings (CI/CD)--check-depsScan package.json dependencies--rules <file>Load custom rules JSON--plugin <file>Load plugin module--compliance owasp-asiKeep only OWASP ASI mapped findings in output"
      },
      {
        "title": "Custom Rules",
        "body": "module.exports = {\n  name: 'my-plugin',\n  patterns: [\n    { id: 'MY_01', cat: 'custom', regex: /dangerous_pattern/g, severity: 'HIGH', desc: 'Description', all: true }\n  ]\n};\n\nguard-scanner ./skills/ --plugin ./my-plugin.js"
      },
      {
        "title": "CI/CD Integration",
        "body": "# .github/workflows/security.yml\n- name: Scan AI skills\n  run: npx -y @guava-parity/guard-scanner ./skills/ --format sarif --fail-on-findings > report.sarif\n- uses: github/codeql-action/upload-sarif@v3\n  with:\n    sarif_file: report.sarif"
      },
      {
        "title": "Threat Categories",
        "body": "35 categories covering OWASP LLM Top 10 + Agentic Security Top 10. See src/patterns.ts for the full pattern database. Key categories:\n\nPrompt Injection — hidden instructions, invisible Unicode, homoglyphs\nIdentity Hijacking ⚿ — persona swap, SOUL.md overwrites, memory wipe\nMemory Poisoning ⚿ — crafted conversation injection\nMCP Security — tool poisoning, SSRF, shadow servers\nA2A Contagion — agent-to-agent worm propagation\nSupply Chain V2 — typosquatting, slopsquatting, lifecycle scripts\nCVE Patterns — CVE-2026-2256, 25046, 25253, 25905, 27825\n\n⚿ = Requires --soul-lock flag"
      }
    ],
    "body": "guard-scanner\n\nScan AI agent skills for 35 categories of threats. v16 adds a 5-layer analysis pipeline, OWASP ASI projection mode, richer finding metadata, and Rust runtime evidence integration on top of the existing prompt injection, identity hijacking, memory poisoning, MCP poisoning, and supply chain coverage.\n\nQuick Start\n# Scan a skill directory\nnpx -y @guava-parity/guard-scanner ./my-skills/ --verbose\n\n# Scan with identity protection\nnpx -y @guava-parity/guard-scanner ./skills/ --soul-lock --strict\n\n# Filter to OWASP ASI mapped findings only\nnpx -y @guava-parity/guard-scanner ./skills/ --compliance owasp-asi --format json\n\n# Installed CLI\nguard-scanner ./skills/ --strict\n\n# npm exec compatibility\nnpm exec --yes --package=@guava-parity/guard-scanner -- guard-scanner ./skills/ --strict\n\nCore Commands\nScan\nguard-scanner <dir>         # Scan directory\nguard-scanner <dir> -v      # Verbose output\nguard-scanner <dir> --json  # JSON report file\nguard-scanner <dir> --sarif # SARIF for CI/CD\nguard-scanner <dir> --html  # HTML report\nguard-scanner <dir> --compliance owasp-asi --format json\n\nAsset Audit\n\nAudit public registries for credential exposure.\n\nguard-scanner audit npm <username>\nguard-scanner audit github <username>\nguard-scanner audit clawhub <query>\nguard-scanner audit all <username> --verbose\n\nMCP Server\n\nStart as MCP server for IDE integration.\n\nguard-scanner serve\n\n\nEditor config (Cursor, Windsurf, Claude Code, OpenClaw):\n\n{\n  \"mcpServers\": {\n    \"guard-scanner\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@guava-parity/guard-scanner\", \"serve\"]\n    }\n  }\n}\n\n\nMCP tools: scan_skill, scan_text, check_tool_call, audit_assets, get_stats, and the async experimental task helpers.\n\nQuality Contract\n\nPublic quality contract:\n\nBenchmark corpus version: 2026-03-13.quality-v1\nPrecision target: >= 0.90\nRecall target: >= 0.90\nFPR/FNR budgets: <= 0.10\nExplainability completeness: 1.0\nRuntime policy latency budget: 5ms\n\nEvidence surfaces:\n\ndocs/spec/capabilities.json\ndocs/data/corpus-metrics.json\ndocs/data/benchmark-ledger.json\ndocs/data/fp-ledger.json\nWatch Mode\n\nMonitor skill directories in real-time during development.\n\nguard-scanner watch ./skills/ --strict --soul-lock\n\nVirusTotal Integration\n\nCombine semantic detection with VirusTotal's 70+ antivirus engines. Optional — guard-scanner works fully without it.\n\nexport VT_API_KEY=your-key\nguard-scanner scan ./skills/ --vt-scan\n\nRuntime Guard\n\nThe validated OpenClaw surface is the compiled runtime plugin entry (dist/openclaw-plugin.mjs) discovered through package.json > openclaw.extensions and mounted on before_tool_call for OpenClaw v2026.3.13, with regression coverage kept on v2026.3.8.\n\nThe before_tool_call hook provides 27 runtime checks across 5 defense layers, while v16 scan output adds a second 5-layer analysis view:\n\nLayer\tFocus\n1. Threat Detection\tReverse shell, curl|bash, SSRF\n2. Trust Defense\tSOUL.md tampering, memory injection\n3. Safety Judge\tPrompt injection in tool arguments\n4. Behavioral\tNo-research execution detection\n5. Trust Exploitation\tAuthority claims, creator bypass\n\nModes: monitor (log only), enforce (block CRITICAL, default), strict (block HIGH+).\n\nv16 Output Surface\nFinding fields: layer, layer_name, owasp_asi, protocol_surface\nCompliance mode: --compliance owasp-asi\nMCP summaries: scan_skill, scan_text, and get_stats now surface layer and ASI context\nRuntime evidence: Rust memory_integrity and soul_hard_gate modules are represented in the TypeScript pipeline\nKey Flags\nFlag\tEffect\n--verbose / -v\tDetailed findings with line numbers\n--strict\tLower detection thresholds\n--soul-lock\tEnable identity protection patterns\n--json / --sarif / --html\tOutput format\n--fail-on-findings\tExit 1 on findings (CI/CD)\n--check-deps\tScan package.json dependencies\n--rules <file>\tLoad custom rules JSON\n--plugin <file>\tLoad plugin module\n--compliance owasp-asi\tKeep only OWASP ASI mapped findings in output\nCustom Rules\nmodule.exports = {\n  name: 'my-plugin',\n  patterns: [\n    { id: 'MY_01', cat: 'custom', regex: /dangerous_pattern/g, severity: 'HIGH', desc: 'Description', all: true }\n  ]\n};\n\nguard-scanner ./skills/ --plugin ./my-plugin.js\n\nCI/CD Integration\n# .github/workflows/security.yml\n- name: Scan AI skills\n  run: npx -y @guava-parity/guard-scanner ./skills/ --format sarif --fail-on-findings > report.sarif\n- uses: github/codeql-action/upload-sarif@v3\n  with:\n    sarif_file: report.sarif\n\nThreat Categories\n\n35 categories covering OWASP LLM Top 10 + Agentic Security Top 10. See src/patterns.ts for the full pattern database. Key categories:\n\nPrompt Injection — hidden instructions, invisible Unicode, homoglyphs\nIdentity Hijacking ⚿ — persona swap, SOUL.md overwrites, memory wipe\nMemory Poisoning ⚿ — crafted conversation injection\nMCP Security — tool poisoning, SSRF, shadow servers\nA2A Contagion — agent-to-agent worm propagation\nSupply Chain V2 — typosquatting, slopsquatting, lifecycle scripts\nCVE Patterns — CVE-2026-2256, 25046, 25253, 25905, 27825\n\n⚿ = Requires --soul-lock flag"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/koatora20/guard-scanner",
    "publisherUrl": "https://clawhub.ai/koatora20/guard-scanner",
    "owner": "koatora20",
    "version": "16.0.1",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/guard-scanner",
    "downloadUrl": "https://openagent3.xyz/downloads/guard-scanner",
    "agentUrl": "https://openagent3.xyz/skills/guard-scanner/agent",
    "manifestUrl": "https://openagent3.xyz/skills/guard-scanner/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/guard-scanner/agent.md"
  }
}