{
  "schemaVersion": "1.0",
  "item": {
    "slug": "external-ki-integration-backup",
    "name": "External Ki Integration Backup",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/konscious0beast/external-ki-integration-backup",
    "canonicalUrl": "https://clawhub.ai/konscious0beast/external-ki-integration-backup",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/external-ki-integration-backup",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=external-ki-integration-backup",
    "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",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-30T16:55:25.780Z",
      "expiresAt": "2026-05-07T16:55:25.780Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network",
        "contentDisposition": "attachment; filename=\"network-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/external-ki-integration-backup"
    },
    "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/external-ki-integration-backup",
    "agentPageUrl": "https://openagent3.xyz/skills/external-ki-integration-backup/agent",
    "manifestUrl": "https://openagent3.xyz/skills/external-ki-integration-backup/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/external-ki-integration-backup/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": "External KI Integration",
        "body": "Use external AI services via browser automation (ChatGPT, Claude, web‑based LLMs) and APIs (Hugging Face Inference, OpenAI‑compatible endpoints) to augment your capabilities."
      },
      {
        "title": "When to use this skill",
        "body": "You need to consult an external AI model (ChatGPT, Claude, Gemini, etc.) for reasoning, analysis, or generation tasks.\nThe user has granted access to their chat interfaces (e.g., via Chrome Relay attached tab).\nYou want to use Hugging Face Inference API (if token provided) for model inference.\nYou need to interact with a free AI demo or Space via browser automation.\nThe task benefits from a second opinion or specialized model (coding, creative writing, summarization)."
      },
      {
        "title": "Requirements",
        "body": "Browser automation – the browser tool with profile=\"chrome\" (user must have attached a tab to OpenClaw Browser Relay).\nExternal AI accounts – user must be logged into the target service (ChatGPT, Claude, etc.) in the attached Chrome tab.\nHugging Face token (optional) – for Inference API access, stored in ~/.openclaw/openclaw.json or provided as environment variable.\nOther API keys (optional) – e.g., OpenAI, Anthropic, if user provides them."
      },
      {
        "title": "Chrome Relay Attachment",
        "body": "The user must click the OpenClaw Browser Relay toolbar icon on the desired tab (badge ON). Verify attachment:\n\nopenclaw browser status\n\nOr via browser tool: browser(action=status, profile=\"chrome\")."
      },
      {
        "title": "Hugging Face Token",
        "body": "If token already stored in config, it will be used automatically. Otherwise, ask user to provide it."
      },
      {
        "title": "Environment Variables (optional)",
        "body": "For API‑based access, you may set:\n\nexport OPENAI_API_KEY=\"sk-...\"\nexport ANTHROPIC_API_KEY=\"sk-ant-...\"\nexport HF_TOKEN=\"hf_...\""
      },
      {
        "title": "General Pattern",
        "body": "Navigate to the service URL (e.g., https://chat.openai.com, https://claude.ai, https://gemini.google.com).\nWait for page load, snapshot with refs=\"aria\" to locate UI elements.\nFind input area (role=\"textbox\", role=\"textbox\" with name \"Message\", etc.).\nType your query using act with ref or selector.\nClick send/submit button (role=\"button\", name=\"Send\").\nWait for response (poll for new text elements, detect loading indicator disappearance).\nExtract response from the output container (role=\"article\", class \"markdown\", etc.).\nReturn the extracted text."
      },
      {
        "title": "Example: ChatGPT via Chrome Relay",
        "body": "// 1. Navigate\nbrowser(action=\"open\", profile=\"chrome\", targetUrl=\"https://chat.openai.com\");\n\n// 2. Snapshot after load\nconst snap = browser(action=\"snapshot\", profile=\"chrome\", refs=\"aria\", interactive=true);\n\n// 3. Find textbox (adapt ref based on snapshot)\nbrowser(action=\"act\", profile=\"chrome\", request={ kind: \"type\", ref: \"textbox:Message\", text: \"Your query here\" });\n\n// 4. Click send button\nbrowser(action=\"act\", profile=\"chrome\", request={ kind: \"click\", ref: \"button:Send\" });\n\n// 5. Wait for response (poll until new text appears)\n// 6. Extract response"
      },
      {
        "title": "Adaptation Notes",
        "body": "UI changes frequently: Use refs=\"aria\" for stable references (aria‑role, aria‑name). Fall back to selector with CSS classes if needed.\nRate limiting: Be gentle; wait 2–5 seconds between interactions.\nSession persistence: The attached tab retains login state; you can continue conversation in same chat."
      },
      {
        "title": "Hugging Face Inference API",
        "body": "See the dedicated Hugging Face skill for detailed usage."
      },
      {
        "title": "OpenAI‑compatible endpoints",
        "body": "If user provides an API key, you can call models via curl or exec:\n\ncurl -s -X POST https://api.openai.com/v1/chat/completions \\\n  -H \"Authorization: Bearer $OPENAI_API_KEY\" \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\n    \"model\": \"gpt-4\",\n    \"messages\": [{\"role\": \"user\", \"content\": \"Hello\"}]\n  }'"
      },
      {
        "title": "Anthropic Claude",
        "body": "curl -s -X POST https://api.anthropic.com/v1/messages \\\n  -H \"x-api-key: $ANTHROPIC_API_KEY\" \\\n  -H \"anthropic-version: 2023-06-01\" \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\n    \"model\": \"claude-3-opus-20240229\",\n    \"max_tokens\": 1024,\n    \"messages\": [{\"role\": \"user\", \"content\": \"Hello\"}]\n  }'"
      },
      {
        "title": "Browser Automation (free)",
        "body": "No direct cost, but uses user's existing subscription (if any).\nRespect rate limits; do not spam requests.\nDo not expose user credentials; rely on attached logged‑in session."
      },
      {
        "title": "API Usage (paid)",
        "body": "Hugging Face Inference: Track estimated costs via system/logs/hf-costs.log. Stay within monthly budget (e.g., 33€). Notify user at 50% threshold.\nOpenAI/Anthropic: If user provides API key, assume they accept associated costs. Still estimate token usage and log if possible.\nGeneral rule: Prefer browser automation for free services; use paid APIs only when explicitly authorized and task justifies cost."
      },
      {
        "title": "Safety",
        "body": "No sensitive data: Avoid sending personal, financial, or confidential information to external services unless user explicitly approves.\nCompliance: Follow external service terms of service.\nFallback: If external service fails, continue with internal reasoning; do not block task completion."
      },
      {
        "title": "Integration with OpenClaw Skills",
        "body": "This skill complements:\n\nHugging Face skill – for dedicated Hugging Face API/Spaces.\nBrowser automation patterns – for generic web interaction.\nMulti‑model orchestration – for delegating sub‑tasks to external models.\n\nAdd this skill to skills/index.md:\n\n| External KI Integration | skills/external‑ki‑integration/SKILL.md |"
      },
      {
        "title": "Example Workflow",
        "body": "Task: Need to generate a complex code snippet.\nCheck: User has ChatGPT tab attached via Chrome Relay.\nOpen ChatGPT, snapshot, locate input.\nType: \"Write a Python function that validates email addresses with regex and DNS MX check.\"\nClick Send.\nWait for response, extract code.\nReturn code to user, optionally refine via follow‑up.\nLog the interaction in memory (pattern learned)."
      },
      {
        "title": "Troubleshooting",
        "body": "Tab not attached: Ask user to click Browser Relay icon on the target tab.\nUI changes: Update refs/selectors based on snapshot.\nRate limits: Wait longer between requests.\nAPI errors: Check token permissions, budget, network."
      },
      {
        "title": "References",
        "body": "OpenClaw Browser Relay docs\nHugging Face skill\nBrowser automation playbook\nUsing free AI models online pattern"
      }
    ],
    "body": "External KI Integration\n\nUse external AI services via browser automation (ChatGPT, Claude, web‑based LLMs) and APIs (Hugging Face Inference, OpenAI‑compatible endpoints) to augment your capabilities.\n\nWhen to use this skill\nYou need to consult an external AI model (ChatGPT, Claude, Gemini, etc.) for reasoning, analysis, or generation tasks.\nThe user has granted access to their chat interfaces (e.g., via Chrome Relay attached tab).\nYou want to use Hugging Face Inference API (if token provided) for model inference.\nYou need to interact with a free AI demo or Space via browser automation.\nThe task benefits from a second opinion or specialized model (coding, creative writing, summarization).\nRequirements\nBrowser automation – the browser tool with profile=\"chrome\" (user must have attached a tab to OpenClaw Browser Relay).\nExternal AI accounts – user must be logged into the target service (ChatGPT, Claude, etc.) in the attached Chrome tab.\nHugging Face token (optional) – for Inference API access, stored in ~/.openclaw/openclaw.json or provided as environment variable.\nOther API keys (optional) – e.g., OpenAI, Anthropic, if user provides them.\nSetup\nChrome Relay Attachment\n\nThe user must click the OpenClaw Browser Relay toolbar icon on the desired tab (badge ON). Verify attachment:\n\nopenclaw browser status\n\n\nOr via browser tool: browser(action=status, profile=\"chrome\").\n\nHugging Face Token\n\nIf token already stored in config, it will be used automatically. Otherwise, ask user to provide it.\n\nEnvironment Variables (optional)\n\nFor API‑based access, you may set:\n\nexport OPENAI_API_KEY=\"sk-...\"\nexport ANTHROPIC_API_KEY=\"sk-ant-...\"\nexport HF_TOKEN=\"hf_...\"\n\nBrowser Automation for Web UIs\nGeneral Pattern\nNavigate to the service URL (e.g., https://chat.openai.com, https://claude.ai, https://gemini.google.com).\nWait for page load, snapshot with refs=\"aria\" to locate UI elements.\nFind input area (role=\"textbox\", role=\"textbox\" with name \"Message\", etc.).\nType your query using act with ref or selector.\nClick send/submit button (role=\"button\", name=\"Send\").\nWait for response (poll for new text elements, detect loading indicator disappearance).\nExtract response from the output container (role=\"article\", class \"markdown\", etc.).\nReturn the extracted text.\nExample: ChatGPT via Chrome Relay\n// 1. Navigate\nbrowser(action=\"open\", profile=\"chrome\", targetUrl=\"https://chat.openai.com\");\n\n// 2. Snapshot after load\nconst snap = browser(action=\"snapshot\", profile=\"chrome\", refs=\"aria\", interactive=true);\n\n// 3. Find textbox (adapt ref based on snapshot)\nbrowser(action=\"act\", profile=\"chrome\", request={ kind: \"type\", ref: \"textbox:Message\", text: \"Your query here\" });\n\n// 4. Click send button\nbrowser(action=\"act\", profile=\"chrome\", request={ kind: \"click\", ref: \"button:Send\" });\n\n// 5. Wait for response (poll until new text appears)\n// 6. Extract response\n\nAdaptation Notes\nUI changes frequently: Use refs=\"aria\" for stable references (aria‑role, aria‑name). Fall back to selector with CSS classes if needed.\nRate limiting: Be gentle; wait 2–5 seconds between interactions.\nSession persistence: The attached tab retains login state; you can continue conversation in same chat.\nAPI Integration\nHugging Face Inference API\n\nSee the dedicated Hugging Face skill for detailed usage.\n\nOpenAI‑compatible endpoints\n\nIf user provides an API key, you can call models via curl or exec:\n\ncurl -s -X POST https://api.openai.com/v1/chat/completions \\\n  -H \"Authorization: Bearer $OPENAI_API_KEY\" \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\n    \"model\": \"gpt-4\",\n    \"messages\": [{\"role\": \"user\", \"content\": \"Hello\"}]\n  }'\n\nAnthropic Claude\ncurl -s -X POST https://api.anthropic.com/v1/messages \\\n  -H \"x-api-key: $ANTHROPIC_API_KEY\" \\\n  -H \"anthropic-version: 2023-06-01\" \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\n    \"model\": \"claude-3-opus-20240229\",\n    \"max_tokens\": 1024,\n    \"messages\": [{\"role\": \"user\", \"content\": \"Hello\"}]\n  }'\n\nCost & Safety\nBrowser Automation (free)\nNo direct cost, but uses user's existing subscription (if any).\nRespect rate limits; do not spam requests.\nDo not expose user credentials; rely on attached logged‑in session.\nAPI Usage (paid)\nHugging Face Inference: Track estimated costs via system/logs/hf-costs.log. Stay within monthly budget (e.g., 33€). Notify user at 50% threshold.\nOpenAI/Anthropic: If user provides API key, assume they accept associated costs. Still estimate token usage and log if possible.\nGeneral rule: Prefer browser automation for free services; use paid APIs only when explicitly authorized and task justifies cost.\nSafety\nNo sensitive data: Avoid sending personal, financial, or confidential information to external services unless user explicitly approves.\nCompliance: Follow external service terms of service.\nFallback: If external service fails, continue with internal reasoning; do not block task completion.\nIntegration with OpenClaw Skills\n\nThis skill complements:\n\nHugging Face skill – for dedicated Hugging Face API/Spaces.\nBrowser automation patterns – for generic web interaction.\nMulti‑model orchestration – for delegating sub‑tasks to external models.\n\nAdd this skill to skills/index.md:\n\n| External KI Integration | skills/external‑ki‑integration/SKILL.md |\n\nExample Workflow\nTask: Need to generate a complex code snippet.\nCheck: User has ChatGPT tab attached via Chrome Relay.\nOpen ChatGPT, snapshot, locate input.\nType: \"Write a Python function that validates email addresses with regex and DNS MX check.\"\nClick Send.\nWait for response, extract code.\nReturn code to user, optionally refine via follow‑up.\nLog the interaction in memory (pattern learned).\nTroubleshooting\nTab not attached: Ask user to click Browser Relay icon on the target tab.\nUI changes: Update refs/selectors based on snapshot.\nRate limits: Wait longer between requests.\nAPI errors: Check token permissions, budget, network.\nReferences\nOpenClaw Browser Relay docs\nHugging Face skill\nBrowser automation playbook\nUsing free AI models online pattern"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/konscious0beast/external-ki-integration-backup",
    "publisherUrl": "https://clawhub.ai/konscious0beast/external-ki-integration-backup",
    "owner": "konscious0beast",
    "version": "1.0.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/external-ki-integration-backup",
    "downloadUrl": "https://openagent3.xyz/downloads/external-ki-integration-backup",
    "agentUrl": "https://openagent3.xyz/skills/external-ki-integration-backup/agent",
    "manifestUrl": "https://openagent3.xyz/skills/external-ki-integration-backup/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/external-ki-integration-backup/agent.md"
  }
}