{
  "schemaVersion": "1.0",
  "item": {
    "slug": "vigil",
    "name": "Vigil",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/RobinOppenstam/vigil",
    "canonicalUrl": "https://clawhub.ai/RobinOppenstam/vigil",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/vigil",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=vigil",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md",
      "_meta.json",
      "scripts/vigil-check.js"
    ],
    "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": "vigil",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-10T16:23:07.052Z",
      "expiresAt": "2026-05-17T16:23:07.052Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=vigil",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=vigil",
        "contentDisposition": "attachment; filename=\"vigil-0.1.1.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "vigil"
      },
      "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/vigil"
    },
    "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/vigil",
    "agentPageUrl": "https://openagent3.xyz/skills/vigil/agent",
    "manifestUrl": "https://openagent3.xyz/skills/vigil/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/vigil/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": "Vigil — Agent Safety Guardrails",
        "body": "Validates what AI agents DO, not what they SAY. Drop-in safety layer for any tool-calling agent."
      },
      {
        "title": "Prerequisites",
        "body": "This skill requires the vigil-agent-safety npm package (12.3KB, Apache 2.0 license):\n\nnpm install vigil-agent-safety\n\nSource code: https://github.com/hexitlabs/vigil\nnpm: https://www.npmjs.com/package/vigil-agent-safety\nThe npm package has zero runtime dependencies. This skill is a wrapper that calls that package."
      },
      {
        "title": "Quick Start",
        "body": "import { checkAction } from 'vigil-agent-safety';\n\nconst result = checkAction({\n  agent: 'my-agent',\n  tool: 'exec',\n  params: { command: 'rm -rf /' },\n});\n\n// result.decision === \"BLOCK\"\n// result.reason === \"Destructive command pattern\"\n// result.latencyMs === 0.3"
      },
      {
        "title": "What It Catches",
        "body": "Destructive commands (rm -rf, mkfs, reverse shells) → BLOCK\nSSRF (metadata endpoints, localhost, internal IPs) → BLOCK\nData exfiltration (curl to external, .ssh/id_rsa access) → BLOCK\nSQL injection (DROP TABLE, UNION SELECT) → BLOCK\nPath traversal (../../../etc/shadow) → BLOCK\nPrompt injection (ignore instructions, [INST] tags) → BLOCK\nEncoding attacks (base64 decode, eval(atob())) → BLOCK\nCredential leaks (API keys, AWS keys, tokens) → ESCALATE\n\n22 rules. Zero dependencies. Under 2ms per check."
      },
      {
        "title": "Modes",
        "body": "import { configure } from 'vigil-agent-safety';\n\n// warn = log violations but don't block (recommended to start)\nconfigure({ mode: 'warn' });\n\n// enforce = block dangerous calls\nconfigure({ mode: 'enforce' });\n\n// log = silent logging only\nconfigure({ mode: 'log' });"
      },
      {
        "title": "Use with Clawdbot",
        "body": "Add Vigil as a safety layer for your agent tool calls. The scripts/vigil-check.js wrapper lets you validate from the command line:\n\n# Check a tool call\nnode scripts/vigil-check.js exec '{\"command\":\"rm -rf /\"}'\n# → BLOCK: Destructive command pattern\n\n# Check a safe call\nnode scripts/vigil-check.js read '{\"path\":\"./README.md\"}'\n# → ALLOW"
      },
      {
        "title": "Policies",
        "body": "Load built-in policy templates:\n\nimport { loadPolicy } from 'vigil-agent-safety';\n\nloadPolicy('restrictive');  // Tightest rules\nloadPolicy('moderate');     // Balanced (default)\nloadPolicy('permissive');   // Minimal blocking"
      },
      {
        "title": "CLI",
        "body": "npx vigil-agent-safety check --tool exec --params '{\"command\":\"ls -la\"}'\nnpx vigil-agent-safety policies"
      },
      {
        "title": "Links",
        "body": "GitHub: https://github.com/hexitlabs/vigil\nnpm: https://www.npmjs.com/package/vigil-agent-safety\nDocs: https://hexitlabs.com/vigil"
      }
    ],
    "body": "Vigil — Agent Safety Guardrails\n\nValidates what AI agents DO, not what they SAY. Drop-in safety layer for any tool-calling agent.\n\nPrerequisites\n\nThis skill requires the vigil-agent-safety npm package (12.3KB, Apache 2.0 license):\n\nnpm install vigil-agent-safety\n\nSource code: https://github.com/hexitlabs/vigil\nnpm: https://www.npmjs.com/package/vigil-agent-safety\nThe npm package has zero runtime dependencies. This skill is a wrapper that calls that package.\nQuick Start\nimport { checkAction } from 'vigil-agent-safety';\n\nconst result = checkAction({\n  agent: 'my-agent',\n  tool: 'exec',\n  params: { command: 'rm -rf /' },\n});\n\n// result.decision === \"BLOCK\"\n// result.reason === \"Destructive command pattern\"\n// result.latencyMs === 0.3\n\nWhat It Catches\nDestructive commands (rm -rf, mkfs, reverse shells) → BLOCK\nSSRF (metadata endpoints, localhost, internal IPs) → BLOCK\nData exfiltration (curl to external, .ssh/id_rsa access) → BLOCK\nSQL injection (DROP TABLE, UNION SELECT) → BLOCK\nPath traversal (../../../etc/shadow) → BLOCK\nPrompt injection (ignore instructions, [INST] tags) → BLOCK\nEncoding attacks (base64 decode, eval(atob())) → BLOCK\nCredential leaks (API keys, AWS keys, tokens) → ESCALATE\n\n22 rules. Zero dependencies. Under 2ms per check.\n\nModes\nimport { configure } from 'vigil-agent-safety';\n\n// warn = log violations but don't block (recommended to start)\nconfigure({ mode: 'warn' });\n\n// enforce = block dangerous calls\nconfigure({ mode: 'enforce' });\n\n// log = silent logging only\nconfigure({ mode: 'log' });\n\nUse with Clawdbot\n\nAdd Vigil as a safety layer for your agent tool calls. The scripts/vigil-check.js wrapper lets you validate from the command line:\n\n# Check a tool call\nnode scripts/vigil-check.js exec '{\"command\":\"rm -rf /\"}'\n# → BLOCK: Destructive command pattern\n\n# Check a safe call\nnode scripts/vigil-check.js read '{\"path\":\"./README.md\"}'\n# → ALLOW\n\nPolicies\n\nLoad built-in policy templates:\n\nimport { loadPolicy } from 'vigil-agent-safety';\n\nloadPolicy('restrictive');  // Tightest rules\nloadPolicy('moderate');     // Balanced (default)\nloadPolicy('permissive');   // Minimal blocking\n\nCLI\nnpx vigil-agent-safety check --tool exec --params '{\"command\":\"ls -la\"}'\nnpx vigil-agent-safety policies\n\nLinks\nGitHub: https://github.com/hexitlabs/vigil\nnpm: https://www.npmjs.com/package/vigil-agent-safety\nDocs: https://hexitlabs.com/vigil"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/RobinOppenstam/vigil",
    "publisherUrl": "https://clawhub.ai/RobinOppenstam/vigil",
    "owner": "RobinOppenstam",
    "version": "0.1.1",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/vigil",
    "downloadUrl": "https://openagent3.xyz/downloads/vigil",
    "agentUrl": "https://openagent3.xyz/skills/vigil/agent",
    "manifestUrl": "https://openagent3.xyz/skills/vigil/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/vigil/agent.md"
  }
}