{
  "schemaVersion": "1.0",
  "item": {
    "slug": "clawshell",
    "name": "Clawshell",
    "source": "tencent",
    "type": "skill",
    "category": "安全合规",
    "sourceUrl": "https://clawhub.ai/polucas/clawshell",
    "canonicalUrl": "https://clawhub.ai/polucas/clawshell",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/clawshell",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=clawshell",
    "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": "clawshell",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-29T04:23:23.321Z",
      "expiresAt": "2026-05-06T04:23:23.321Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=clawshell",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=clawshell",
        "contentDisposition": "attachment; filename=\"clawshell-0.1.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "clawshell"
      },
      "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/clawshell"
    },
    "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/clawshell",
    "agentPageUrl": "https://openagent3.xyz/skills/clawshell/agent",
    "manifestUrl": "https://openagent3.xyz/skills/clawshell/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/clawshell/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": "ClawShell",
        "body": "Human-in-the-loop security layer for OpenClaw. ClawShell intercepts shell commands before execution, analyzes their risk level, and requires your explicit approval (via push notification) for dangerous operations."
      },
      {
        "title": "How it works",
        "body": "The agent calls clawshell_bash instead of bash\nClawShell analyzes the command against built-in and configurable risk rules\nBased on risk level:\n\nCritical (e.g. rm -rf /, fork bombs) — automatically blocked\nHigh (e.g. rm -rf, curl to external URLs, credential access) — sends a push notification and waits for your approval\nMedium (e.g. npm install, git push) — logged and allowed\nLow (e.g. ls, cat, git status) — allowed\n\n\nAll decisions are logged to logs/clawshell.jsonl"
      },
      {
        "title": "clawshell_bash",
        "body": "Secure replacement for bash. Analyzes command risk and executes only if safe or approved.\n\nParameters:\n\ncommand (string, required) — The shell command to execute\nworkingDir (string, optional) — Working directory (defaults to cwd)\n\nReturns: { exitCode, stdout, stderr }\n\nHigh-risk commands will block until you approve or reject via push notification. Critical commands are rejected immediately."
      },
      {
        "title": "clawshell_status",
        "body": "Returns current ClawShell state: pending approval requests and recent decisions.\n\nParameters: none"
      },
      {
        "title": "clawshell_logs",
        "body": "Returns recent log entries for audit and debugging.\n\nParameters:\n\ncount (number, optional) — Number of entries to return (default: 20)"
      },
      {
        "title": "1. Install dependencies",
        "body": "cd /app/workspace/skills/clawshell\nnpm install"
      },
      {
        "title": "2. Configure Pushover notifications",
        "body": "Create a Pushover application at https://pushover.net/apps/build and add your keys to .env:\n\nCLAWSHELL_PUSHOVER_USER=your-user-key\nCLAWSHELL_PUSHOVER_TOKEN=your-app-token\n\nAlternatively, configure Telegram instead:\n\nCLAWSHELL_TELEGRAM_BOT_TOKEN=your-bot-token\nCLAWSHELL_TELEGRAM_CHAT_ID=your-chat-id"
      },
      {
        "title": "3. Add to TOOLS.md",
        "body": "Add the following to your OpenClaw TOOLS.md so the agent uses ClawShell for shell commands:\n\n## Shell Access\n\nUse `clawshell_bash` for ALL shell command execution. Do not use `bash` directly.\nClawShell will analyze commands for risk and require human approval for dangerous operations.\n\nAvailable tools:\n- `clawshell_bash(command, workingDir)` — Execute a shell command with risk analysis\n- `clawshell_status()` — Check pending approvals and recent decisions\n- `clawshell_logs(count)` — View recent audit log entries"
      },
      {
        "title": "Configuration",
        "body": "ClawShell reads configuration from environment variables (CLAWSHELL_*) with fallback to config.yaml.\n\nVariableDefaultDescriptionCLAWSHELL_PUSHOVER_USER—Pushover user keyCLAWSHELL_PUSHOVER_TOKEN—Pushover app tokenCLAWSHELL_TELEGRAM_BOT_TOKEN—Telegram bot token (alternative)CLAWSHELL_TELEGRAM_CHAT_ID—Telegram chat ID (alternative)CLAWSHELL_TIMEOUT_SECONDS300Seconds to wait for approval before auto-rejectCLAWSHELL_LOG_DIRlogs/Directory for JSONL log filesCLAWSHELL_LOG_LEVELinfoLog verbosity: debug, info, warn, errorCLAWSHELL_BLOCKLIST—Comma-separated extra blocked commandsCLAWSHELL_ALLOWLIST—Comma-separated extra allowed commands\n\nCustom rules can also be defined in config.yaml under rules.blocklist and rules.allowlist using exact strings, globs, or regex patterns."
      },
      {
        "title": "Limitations",
        "body": "Not a security guarantee. LLMs can encode, split, or obfuscate commands to bypass pattern matching.\nDefense-in-depth only. Use alongside OpenClaw's sandbox mode, not as a replacement.\nApproval latency. High-risk commands block execution until you respond or the timeout expires.\n\nAlways ask your AI to scan any skill or software for security risks."
      }
    ],
    "body": "ClawShell\n\nHuman-in-the-loop security layer for OpenClaw. ClawShell intercepts shell commands before execution, analyzes their risk level, and requires your explicit approval (via push notification) for dangerous operations.\n\nHow it works\nThe agent calls clawshell_bash instead of bash\nClawShell analyzes the command against built-in and configurable risk rules\nBased on risk level:\nCritical (e.g. rm -rf /, fork bombs) — automatically blocked\nHigh (e.g. rm -rf, curl to external URLs, credential access) — sends a push notification and waits for your approval\nMedium (e.g. npm install, git push) — logged and allowed\nLow (e.g. ls, cat, git status) — allowed\nAll decisions are logged to logs/clawshell.jsonl\nTools\nclawshell_bash\n\nSecure replacement for bash. Analyzes command risk and executes only if safe or approved.\n\nParameters:\n\ncommand (string, required) — The shell command to execute\nworkingDir (string, optional) — Working directory (defaults to cwd)\n\nReturns: { exitCode, stdout, stderr }\n\nHigh-risk commands will block until you approve or reject via push notification. Critical commands are rejected immediately.\n\nclawshell_status\n\nReturns current ClawShell state: pending approval requests and recent decisions.\n\nParameters: none\n\nclawshell_logs\n\nReturns recent log entries for audit and debugging.\n\nParameters:\n\ncount (number, optional) — Number of entries to return (default: 20)\nSetup\n1. Install dependencies\ncd /app/workspace/skills/clawshell\nnpm install\n\n2. Configure Pushover notifications\n\nCreate a Pushover application at https://pushover.net/apps/build and add your keys to .env:\n\nCLAWSHELL_PUSHOVER_USER=your-user-key\nCLAWSHELL_PUSHOVER_TOKEN=your-app-token\n\n\nAlternatively, configure Telegram instead:\n\nCLAWSHELL_TELEGRAM_BOT_TOKEN=your-bot-token\nCLAWSHELL_TELEGRAM_CHAT_ID=your-chat-id\n\n3. Add to TOOLS.md\n\nAdd the following to your OpenClaw TOOLS.md so the agent uses ClawShell for shell commands:\n\n## Shell Access\n\nUse `clawshell_bash` for ALL shell command execution. Do not use `bash` directly.\nClawShell will analyze commands for risk and require human approval for dangerous operations.\n\nAvailable tools:\n- `clawshell_bash(command, workingDir)` — Execute a shell command with risk analysis\n- `clawshell_status()` — Check pending approvals and recent decisions\n- `clawshell_logs(count)` — View recent audit log entries\n\nConfiguration\n\nClawShell reads configuration from environment variables (CLAWSHELL_*) with fallback to config.yaml.\n\nVariable\tDefault\tDescription\nCLAWSHELL_PUSHOVER_USER\t—\tPushover user key\nCLAWSHELL_PUSHOVER_TOKEN\t—\tPushover app token\nCLAWSHELL_TELEGRAM_BOT_TOKEN\t—\tTelegram bot token (alternative)\nCLAWSHELL_TELEGRAM_CHAT_ID\t—\tTelegram chat ID (alternative)\nCLAWSHELL_TIMEOUT_SECONDS\t300\tSeconds to wait for approval before auto-reject\nCLAWSHELL_LOG_DIR\tlogs/\tDirectory for JSONL log files\nCLAWSHELL_LOG_LEVEL\tinfo\tLog verbosity: debug, info, warn, error\nCLAWSHELL_BLOCKLIST\t—\tComma-separated extra blocked commands\nCLAWSHELL_ALLOWLIST\t—\tComma-separated extra allowed commands\n\nCustom rules can also be defined in config.yaml under rules.blocklist and rules.allowlist using exact strings, globs, or regex patterns.\n\nLimitations\nNot a security guarantee. LLMs can encode, split, or obfuscate commands to bypass pattern matching.\nDefense-in-depth only. Use alongside OpenClaw's sandbox mode, not as a replacement.\nApproval latency. High-risk commands block execution until you respond or the timeout expires.\n\nAlways ask your AI to scan any skill or software for security risks."
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/polucas/clawshell",
    "publisherUrl": "https://clawhub.ai/polucas/clawshell",
    "owner": "polucas",
    "version": "0.1.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/clawshell",
    "downloadUrl": "https://openagent3.xyz/downloads/clawshell",
    "agentUrl": "https://openagent3.xyz/skills/clawshell/agent",
    "manifestUrl": "https://openagent3.xyz/skills/clawshell/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/clawshell/agent.md"
  }
}