{
  "schemaVersion": "1.0",
  "item": {
    "slug": "agi-terminal-helper",
    "name": "Agi Terminal Helper",
    "source": "tencent",
    "type": "skill",
    "category": "安全合规",
    "sourceUrl": "https://clawhub.ai/liuwujijay/agi-terminal-helper",
    "canonicalUrl": "https://clawhub.ai/liuwujijay/agi-terminal-helper",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/agi-terminal-helper",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=agi-terminal-helper",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md",
      "_meta.json"
    ],
    "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": "agi-terminal-helper",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-24T01:26:40.331Z",
      "expiresAt": "2026-05-01T01:26:40.331Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=agi-terminal-helper",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=agi-terminal-helper",
        "contentDisposition": "attachment; filename=\"agi-terminal-helper-1.0.2.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "agi-terminal-helper"
      },
      "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/agi-terminal-helper"
    },
    "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/agi-terminal-helper",
    "agentPageUrl": "https://openagent3.xyz/skills/agi-terminal-helper/agent",
    "manifestUrl": "https://openagent3.xyz/skills/agi-terminal-helper/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/agi-terminal-helper/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": "Terminal Helper — a runbook for OpenClaw exec",
        "body": "This skill is not a “generic terminal tips” template.\nIt’s a concrete runbook for how to use OpenClaw’s exec tool effectively in a real workspace (like your /Users/.../clawd workspace), with attention to:\n\nsandbox vs host execution\npredictable working directories\nlong-running processes\npermissions on macOS (Peekaboo, screen recording, UI automation)\navoiding “accidental shell scripting” disasters\n\nOpenClaw skills are loaded from bundled skills, ~/.openclaw/skills, and <workspace>/skills with workspace taking precedence. :contentReference[oaicite:12]{index=12}"
      },
      {
        "title": "1) State the intent + the exact command before running it",
        "body": "Before calling exec, I will say:\n\nwhat the command is intended to do\nwhat directory it will run in\nwhat files it might read/write\nwhat output I expect (so we can spot anomalies)"
      },
      {
        "title": "2) Default to read-only exploration",
        "body": "When debugging or orienting:\n\npwd, ls -la, git status, rg, cat, head, tail\nonly escalate to writes/installs after we know what’s going on"
      },
      {
        "title": "3) Prefer sandboxed execution for untrusted or high-churn work",
        "body": "Use the sandbox for:\n\ntests, builds, dependency installs\nexploring unknown repos\nrunning scripts from third-party sources\n\nImportant nuance:\nIf a session is sandboxed, the sandbox does not inherit host process.env.\nGlobal env and skills.entries.<skill>.env/apiKey apply to host runs only; sandbox env must be set separately. :contentReference[oaicite:13]{index=13}"
      },
      {
        "title": "4) Explicit confirmation for anything risky",
        "body": "I will require the user to confirm before:\n\ndeleting or overwriting files\ninstalling system-level packages\ntouching ~/.ssh, keychains, browser profiles\nchanging network/system settings\nrunning privileged commands (sudo, launchctl changes)"
      },
      {
        "title": "A) Choose a working directory deliberately",
        "body": "When diagnosing OpenClaw itself, I’ll work inside your workspace (example: /Users/proman/clawd) and be explicit about it.\n\nTypical commands:\n\ncheck skills:\n\nls -la ./skills\nfind ./skills -maxdepth 2 -name SKILL.md -print\n\n\ncheck git state:\n\ngit status (if the workspace is a git repo)\n\n\nverify binaries:\n\nwhich peekaboo || echo \"peekaboo not on PATH\""
      },
      {
        "title": "B) Keep commands single-purpose",
        "body": "Prefer multiple small commands over one “do everything” pipeline. This makes it easier to review and safer to approve."
      },
      {
        "title": "C) Long-running commands: background + poll",
        "body": "When supported, run with a short yield and then poll a process session.\n\nExamples you can adapt:\n\nstart a long build:\n\nexec: make test (with a short yield)\n\n\npoll until completion:\n\nprocess: poll (using the returned session id)\n\n(Exact parameter names depend on your tool surface, but the pattern is: yield → poll.)"
      },
      {
        "title": "Playbook 1: “My skill isn’t loading”",
        "body": "Confirm skill location/precedence:\n\nOpenClaw loads <workspace>/skills and that wins precedence. :contentReference[oaicite:14]{index=14}\n\n\nVerify the skill folder has SKILL.md and valid frontmatter.\nIf you changed files, ensure watcher is enabled:\n\nskills.load.watch: true is the default pattern. :contentReference[oaicite:15]{index=15}"
      },
      {
        "title": "Playbook 2: “Peekaboo works in Terminal but fails in OpenClaw”",
        "body": "This is usually macOS TCC context + daemon behavior. A common fix is enabling PeekabooBridge in OpenClaw.app:\n\nSettings → Enable Peekaboo Bridge :contentReference[oaicite:16]{index=16}\n\nThen validate:\n\npeekaboo bridge status --verbose should select a host (OpenClaw.app) rather than local (in-process). :contentReference[oaicite:17]{index=17}"
      },
      {
        "title": "Playbook 3: “ClawHub sync rejects my skill docs”",
        "body": "ClawHub has a quality gate (language-aware word counting and heuristics) that rejects docs that are too thin/templated. :contentReference[oaicite:18]{index=18}\nFix by adding:\n\nconcrete examples\ntroubleshooting\nenvironment notes (sandbox, PATH, permissions)\n“what/why/when/how” that is clearly specific to the skill"
      },
      {
        "title": "What I will NOT do",
        "body": "I will not run remote “install scripts” (e.g., curl | sh) without explicit user request and review.\nI will not paste or echo secrets into commands.\nI will not make destructive changes without confirming the exact file paths."
      },
      {
        "title": "Quick commands I often start with",
        "body": "pwd\nls -la\ngit status\nrg -n \"error|warn|TODO\" .\nuname -a\nnode -v && python -V\n\nIf you want raw, direct execution (no model involvement), use /term."
      }
    ],
    "body": "Terminal Helper — a runbook for OpenClaw exec\n\nThis skill is not a “generic terminal tips” template. It’s a concrete runbook for how to use OpenClaw’s exec tool effectively in a real workspace (like your /Users/.../clawd workspace), with attention to:\n\nsandbox vs host execution\npredictable working directories\nlong-running processes\npermissions on macOS (Peekaboo, screen recording, UI automation)\navoiding “accidental shell scripting” disasters\n\nOpenClaw skills are loaded from bundled skills, ~/.openclaw/skills, and <workspace>/skills with workspace taking precedence. :contentReference[oaicite:12]{index=12}\n\nOperating principles (what I will do every time)\n1) State the intent + the exact command before running it\n\nBefore calling exec, I will say:\n\nwhat the command is intended to do\nwhat directory it will run in\nwhat files it might read/write\nwhat output I expect (so we can spot anomalies)\n2) Default to read-only exploration\n\nWhen debugging or orienting:\n\npwd, ls -la, git status, rg, cat, head, tail\nonly escalate to writes/installs after we know what’s going on\n3) Prefer sandboxed execution for untrusted or high-churn work\n\nUse the sandbox for:\n\ntests, builds, dependency installs\nexploring unknown repos\nrunning scripts from third-party sources\n\nImportant nuance: If a session is sandboxed, the sandbox does not inherit host process.env. Global env and skills.entries.<skill>.env/apiKey apply to host runs only; sandbox env must be set separately. :contentReference[oaicite:13]{index=13}\n\n4) Explicit confirmation for anything risky\n\nI will require the user to confirm before:\n\ndeleting or overwriting files\ninstalling system-level packages\ntouching ~/.ssh, keychains, browser profiles\nchanging network/system settings\nrunning privileged commands (sudo, launchctl changes)\nExecution patterns (the “how”)\nA) Choose a working directory deliberately\n\nWhen diagnosing OpenClaw itself, I’ll work inside your workspace (example: /Users/proman/clawd) and be explicit about it.\n\nTypical commands:\n\ncheck skills:\nls -la ./skills\nfind ./skills -maxdepth 2 -name SKILL.md -print\ncheck git state:\ngit status (if the workspace is a git repo)\nverify binaries:\nwhich peekaboo || echo \"peekaboo not on PATH\"\nB) Keep commands single-purpose\n\nPrefer multiple small commands over one “do everything” pipeline. This makes it easier to review and safer to approve.\n\nC) Long-running commands: background + poll\n\nWhen supported, run with a short yield and then poll a process session.\n\nExamples you can adapt:\n\nstart a long build:\nexec: make test (with a short yield)\npoll until completion:\nprocess: poll (using the returned session id)\n\n(Exact parameter names depend on your tool surface, but the pattern is: yield → poll.)\n\nPractical playbooks\nPlaybook 1: “My skill isn’t loading”\nConfirm skill location/precedence:\nOpenClaw loads <workspace>/skills and that wins precedence. :contentReference[oaicite:14]{index=14}\nVerify the skill folder has SKILL.md and valid frontmatter.\nIf you changed files, ensure watcher is enabled:\nskills.load.watch: true is the default pattern. :contentReference[oaicite:15]{index=15}\nPlaybook 2: “Peekaboo works in Terminal but fails in OpenClaw”\n\nThis is usually macOS TCC context + daemon behavior. A common fix is enabling PeekabooBridge in OpenClaw.app:\n\nSettings → Enable Peekaboo Bridge :contentReference[oaicite:16]{index=16}\n\nThen validate:\n\npeekaboo bridge status --verbose should select a host (OpenClaw.app) rather than local (in-process). :contentReference[oaicite:17]{index=17}\nPlaybook 3: “ClawHub sync rejects my skill docs”\n\nClawHub has a quality gate (language-aware word counting and heuristics) that rejects docs that are too thin/templated. :contentReference[oaicite:18]{index=18} Fix by adding:\n\nconcrete examples\ntroubleshooting\nenvironment notes (sandbox, PATH, permissions)\n“what/why/when/how” that is clearly specific to the skill\nWhat I will NOT do\nI will not run remote “install scripts” (e.g., curl | sh) without explicit user request and review.\nI will not paste or echo secrets into commands.\nI will not make destructive changes without confirming the exact file paths.\nQuick commands I often start with\npwd\nls -la\ngit status\nrg -n \"error|warn|TODO\" .\nuname -a\nnode -v && python -V\n\nIf you want raw, direct execution (no model involvement), use /term."
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/liuwujijay/agi-terminal-helper",
    "publisherUrl": "https://clawhub.ai/liuwujijay/agi-terminal-helper",
    "owner": "liuwujijay",
    "version": "1.0.2",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/agi-terminal-helper",
    "downloadUrl": "https://openagent3.xyz/downloads/agi-terminal-helper",
    "agentUrl": "https://openagent3.xyz/skills/agi-terminal-helper/agent",
    "manifestUrl": "https://openagent3.xyz/skills/agi-terminal-helper/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/agi-terminal-helper/agent.md"
  }
}