{
  "schemaVersion": "1.0",
  "item": {
    "slug": "arc-skill-sandbox",
    "name": "Skill Sandbox",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/Trypto1019/arc-skill-sandbox",
    "canonicalUrl": "https://clawhub.ai/Trypto1019/arc-skill-sandbox",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/arc-skill-sandbox",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=arc-skill-sandbox",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md",
      "scripts/sandbox.py"
    ],
    "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": "2026-05-07T17:22:31.273Z",
      "expiresAt": "2026-05-14T17:22:31.273Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=afrexai-annual-report",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=afrexai-annual-report",
        "contentDisposition": "attachment; filename=\"afrexai-annual-report-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/arc-skill-sandbox"
    },
    "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/arc-skill-sandbox",
    "agentPageUrl": "https://openagent3.xyz/skills/arc-skill-sandbox/agent",
    "manifestUrl": "https://openagent3.xyz/skills/arc-skill-sandbox/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/arc-skill-sandbox/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": "Skill Sandbox",
        "body": "Run untrusted skills in a monitored environment. See exactly what they do before giving them access to your real system."
      },
      {
        "title": "Why This Exists",
        "body": "ClawHub has hundreds of skills. Some are malicious. Even after scanning with arc-skill-scanner, you can't catch everything with static analysis. The sandbox lets you run a skill's scripts and observe their behavior at runtime — what network calls they make, what files they access, what environment variables they read."
      },
      {
        "title": "Sandbox a skill directory",
        "body": "python3 {baseDir}/scripts/sandbox.py run --path ~/.openclaw/skills/some-skill/"
      },
      {
        "title": "Run a specific script in sandbox",
        "body": "python3 {baseDir}/scripts/sandbox.py run --script ~/.openclaw/skills/some-skill/scripts/main.py"
      },
      {
        "title": "Run with network monitoring",
        "body": "python3 {baseDir}/scripts/sandbox.py run --path ~/.openclaw/skills/some-skill/ --monitor-network"
      },
      {
        "title": "Run with fake environment variables",
        "body": "python3 {baseDir}/scripts/sandbox.py run --path ~/.openclaw/skills/some-skill/ --fake-env"
      },
      {
        "title": "Run with a time limit",
        "body": "python3 {baseDir}/scripts/sandbox.py run --path ~/.openclaw/skills/some-skill/ --timeout 30"
      },
      {
        "title": "Generate a safety report",
        "body": "python3 {baseDir}/scripts/sandbox.py report --path ~/.openclaw/skills/some-skill/"
      },
      {
        "title": "Filesystem Access",
        "body": "Files opened (read/write)\nDirectories created\nFile deletions\nPermission changes"
      },
      {
        "title": "Environment Variables",
        "body": "Which env vars are read\nWhether sensitive keys are accessed (API keys, tokens, passwords)\nOption to inject fake values to see what the skill does with them"
      },
      {
        "title": "Network Activity",
        "body": "Outbound HTTP/HTTPS requests (URLs, methods, payloads)\nDNS lookups\nSocket connections\nFTP, SMTP, and other protocols"
      },
      {
        "title": "Process Execution",
        "body": "Subprocess calls\nShell commands\nDynamic imports"
      },
      {
        "title": "Safety Modes",
        "body": "observe (default) — Run the skill and log everything it does. No restrictions.\nrestricted — Block network access and filesystem writes outside a temp directory.\nhoneypot — Provide fake credentials and endpoints to see if the skill tries to exfiltrate."
      },
      {
        "title": "Output",
        "body": "The sandbox produces a JSON report with:\n\nAll filesystem operations (reads, writes, deletes)\nAll environment variable accesses\nAll network connections attempted\nAll subprocess calls\nWarnings for suspicious patterns\nA safety verdict (SAFE / SUSPICIOUS / DANGEROUS)"
      },
      {
        "title": "Integration",
        "body": "Combine with the workflow orchestrator for automated pre-install checks:\n\nscan skill → sandbox run → review report → install if safe → audit log"
      },
      {
        "title": "Limitations",
        "body": "Python skills only (JavaScript/shell support planned)\nCannot catch all evasion techniques (obfuscated or delayed execution)\nNetwork monitoring requires the skill to use standard Python libraries\nNot a true OS-level sandbox (use Docker for that level of isolation)"
      }
    ],
    "body": "Skill Sandbox\n\nRun untrusted skills in a monitored environment. See exactly what they do before giving them access to your real system.\n\nWhy This Exists\n\nClawHub has hundreds of skills. Some are malicious. Even after scanning with arc-skill-scanner, you can't catch everything with static analysis. The sandbox lets you run a skill's scripts and observe their behavior at runtime — what network calls they make, what files they access, what environment variables they read.\n\nCommands\nSandbox a skill directory\npython3 {baseDir}/scripts/sandbox.py run --path ~/.openclaw/skills/some-skill/\n\nRun a specific script in sandbox\npython3 {baseDir}/scripts/sandbox.py run --script ~/.openclaw/skills/some-skill/scripts/main.py\n\nRun with network monitoring\npython3 {baseDir}/scripts/sandbox.py run --path ~/.openclaw/skills/some-skill/ --monitor-network\n\nRun with fake environment variables\npython3 {baseDir}/scripts/sandbox.py run --path ~/.openclaw/skills/some-skill/ --fake-env\n\nRun with a time limit\npython3 {baseDir}/scripts/sandbox.py run --path ~/.openclaw/skills/some-skill/ --timeout 30\n\nGenerate a safety report\npython3 {baseDir}/scripts/sandbox.py report --path ~/.openclaw/skills/some-skill/\n\nWhat It Monitors\nFilesystem Access\nFiles opened (read/write)\nDirectories created\nFile deletions\nPermission changes\nEnvironment Variables\nWhich env vars are read\nWhether sensitive keys are accessed (API keys, tokens, passwords)\nOption to inject fake values to see what the skill does with them\nNetwork Activity\nOutbound HTTP/HTTPS requests (URLs, methods, payloads)\nDNS lookups\nSocket connections\nFTP, SMTP, and other protocols\nProcess Execution\nSubprocess calls\nShell commands\nDynamic imports\nSafety Modes\nobserve (default) — Run the skill and log everything it does. No restrictions.\nrestricted — Block network access and filesystem writes outside a temp directory.\nhoneypot — Provide fake credentials and endpoints to see if the skill tries to exfiltrate.\nOutput\n\nThe sandbox produces a JSON report with:\n\nAll filesystem operations (reads, writes, deletes)\nAll environment variable accesses\nAll network connections attempted\nAll subprocess calls\nWarnings for suspicious patterns\nA safety verdict (SAFE / SUSPICIOUS / DANGEROUS)\nIntegration\n\nCombine with the workflow orchestrator for automated pre-install checks:\n\nscan skill → sandbox run → review report → install if safe → audit log\n\nLimitations\nPython skills only (JavaScript/shell support planned)\nCannot catch all evasion techniques (obfuscated or delayed execution)\nNetwork monitoring requires the skill to use standard Python libraries\nNot a true OS-level sandbox (use Docker for that level of isolation)"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/Trypto1019/arc-skill-sandbox",
    "publisherUrl": "https://clawhub.ai/Trypto1019/arc-skill-sandbox",
    "owner": "Trypto1019",
    "version": "1.1.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/arc-skill-sandbox",
    "downloadUrl": "https://openagent3.xyz/downloads/arc-skill-sandbox",
    "agentUrl": "https://openagent3.xyz/skills/arc-skill-sandbox/agent",
    "manifestUrl": "https://openagent3.xyz/skills/arc-skill-sandbox/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/arc-skill-sandbox/agent.md"
  }
}