{
  "schemaVersion": "1.0",
  "item": {
    "slug": "hugging-face",
    "name": "Hugging Face",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/ivangdavila/hugging-face",
    "canonicalUrl": "https://clawhub.ai/ivangdavila/hugging-face",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/hugging-face",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=hugging-face",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md",
      "discovery.md",
      "evaluation.md",
      "inference.md",
      "memory-template.md",
      "setup.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/hugging-face"
    },
    "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/hugging-face",
    "agentPageUrl": "https://openagent3.xyz/skills/hugging-face/agent",
    "manifestUrl": "https://openagent3.xyz/skills/hugging-face/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/hugging-face/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": "Setup",
        "body": "On first use, read setup.md for integration guidelines and local memory initialization."
      },
      {
        "title": "When to Use",
        "body": "User needs to find the right Hugging Face model, dataset, or Space for a concrete task and move from browsing to reliable execution.\nAgent handles discovery, filtering, license checks, quick benchmarking, and integration-ready inference plans."
      },
      {
        "title": "Architecture",
        "body": "Memory and reusable artifacts live in ~/hugging-face/. See memory-template.md for structure and status fields.\n\n~/hugging-face/\n|- memory.md          # Stable context, priorities, and defaults\n|- shortlists.md      # Candidate models and datasets by use case\n|- evaluations.md     # Benchmark runs, winners, and caveats\n|- endpoints.md       # Approved endpoints and auth notes\n`- exports/           # Saved outputs and comparison snapshots"
      },
      {
        "title": "Quick Reference",
        "body": "Load only one focused file at a time to keep context small and decisions explicit.\n\nTopicFileSetup processsetup.mdMemory templatememory-template.mdModel and dataset discoverydiscovery.mdInference execution patternsinference.mdEvaluation rubric and scoringevaluation.mdCommon failures and recoverytroubleshooting.md"
      },
      {
        "title": "1. Lock Objective and Constraints First",
        "body": "Before selecting any artifact, confirm task type, latency budget, cost boundary, and deployment target.\n\nUse this minimum scope packet:\n\nTask type: chat, generation, embedding, classification, vision, or speech\nQuality priority: best quality, best speed, or balanced\nRuntime constraints: CPU only, specific GPU class, or hosted endpoint\nCompliance constraints: license, region, or private data limits"
      },
      {
        "title": "2. Separate Discovery from Execution",
        "body": "Do not run inference on the first candidate found.\n\nFirst create a shortlist of at least three candidates, then execute only on finalists that pass compatibility and license checks."
      },
      {
        "title": "3. Validate License and Access Before Recommendation",
        "body": "For every candidate, verify license, gated access status, model size, and framework compatibility.\n\nIf any of these are unknown, mark the candidate as provisional and avoid production recommendation."
      },
      {
        "title": "4. Benchmark with a Deterministic Mini Suite",
        "body": "Use the same prompt set and output checks across candidates so results are comparable.\n\nMinimum benchmark set:\n\nOne typical request\nOne edge-case request\nOne failure-prone request"
      },
      {
        "title": "5. Minimize External Data",
        "body": "Send only what is required for the selected endpoint.\n\nNever send credentials, local paths, or unrelated private context in request payloads."
      },
      {
        "title": "6. Use a Fallback Ladder",
        "body": "If the preferred model fails, apply ordered fallback:\n\nRetry same endpoint with smaller payload\nSwitch to a compatible backup model\nSwitch to local-only workflow if available"
      },
      {
        "title": "7. Keep Runs Reproducible",
        "body": "Log selected model id, endpoint, key parameters, and evaluation result in local memory so future runs are consistent and auditable."
      },
      {
        "title": "Common Traps",
        "body": "Picking the highest download count as the only criterion -> often misses license, latency, or domain fit.\nIgnoring gated model requirements -> integration fails at runtime due to access restrictions.\nComparing models with different prompts -> quality conclusions become unreliable.\nSending full user context to inference endpoints -> unnecessary privacy exposure.\nSkipping fallback design -> workflows fail hard on transient endpoint errors."
      },
      {
        "title": "External Endpoints",
        "body": "Use discovery endpoints before inference so candidate selection remains explainable and reproducible.\n\nEndpointData SentPurposehttps://huggingface.co/api/modelsSearch terms, filter parametersDiscover model candidateshttps://huggingface.co/api/datasetsSearch terms, filter parametersDiscover dataset candidateshttps://huggingface.co/api/spacesSearch terms, filter parametersDiscover runnable Spaceshttps://api-inference.huggingface.co/models/{model_id}Prompt or task input payload, selected model id, auth tokenRun hosted inference\n\nNo other data is sent externally."
      },
      {
        "title": "Security & Privacy",
        "body": "Data that leaves your machine:\n\nSearch terms and filter inputs sent to Hugging Face discovery APIs.\nInference payloads sent to Hugging Face Inference API when execution is requested.\n\nData that stays local:\n\nPreferences, shortlists, evaluation notes, and endpoint decisions in ~/hugging-face/.\n\nThis skill does NOT:\n\nExfiltrate local files by default.\nSend undeclared network requests.\nStore raw secrets in local notes.\nModify its own skill definition file."
      },
      {
        "title": "Trust",
        "body": "By using this skill, selected request data is sent to Hugging Face services.\nOnly install if you trust Hugging Face with the inputs you choose to process."
      },
      {
        "title": "Related Skills",
        "body": "Install with clawhub install <slug> if user confirms:\n\nai - general AI strategy and model-selection framing\napi - API-first integration patterns and HTTP debugging\ndata-analysis - dataset inspection and quality interpretation\ndata - structured data workflows and extraction patterns\ncode - implementation support for scripts and adapters"
      },
      {
        "title": "Feedback",
        "body": "If useful: clawhub star hugging-face\nStay updated: clawhub sync"
      }
    ],
    "body": "Setup\n\nOn first use, read setup.md for integration guidelines and local memory initialization.\n\nWhen to Use\n\nUser needs to find the right Hugging Face model, dataset, or Space for a concrete task and move from browsing to reliable execution. Agent handles discovery, filtering, license checks, quick benchmarking, and integration-ready inference plans.\n\nArchitecture\n\nMemory and reusable artifacts live in ~/hugging-face/. See memory-template.md for structure and status fields.\n\n~/hugging-face/\n|- memory.md          # Stable context, priorities, and defaults\n|- shortlists.md      # Candidate models and datasets by use case\n|- evaluations.md     # Benchmark runs, winners, and caveats\n|- endpoints.md       # Approved endpoints and auth notes\n`- exports/           # Saved outputs and comparison snapshots\n\nQuick Reference\n\nLoad only one focused file at a time to keep context small and decisions explicit.\n\nTopic\tFile\nSetup process\tsetup.md\nMemory template\tmemory-template.md\nModel and dataset discovery\tdiscovery.md\nInference execution patterns\tinference.md\nEvaluation rubric and scoring\tevaluation.md\nCommon failures and recovery\ttroubleshooting.md\nCore Rules\n1. Lock Objective and Constraints First\n\nBefore selecting any artifact, confirm task type, latency budget, cost boundary, and deployment target.\n\nUse this minimum scope packet:\n\nTask type: chat, generation, embedding, classification, vision, or speech\nQuality priority: best quality, best speed, or balanced\nRuntime constraints: CPU only, specific GPU class, or hosted endpoint\nCompliance constraints: license, region, or private data limits\n2. Separate Discovery from Execution\n\nDo not run inference on the first candidate found.\n\nFirst create a shortlist of at least three candidates, then execute only on finalists that pass compatibility and license checks.\n\n3. Validate License and Access Before Recommendation\n\nFor every candidate, verify license, gated access status, model size, and framework compatibility.\n\nIf any of these are unknown, mark the candidate as provisional and avoid production recommendation.\n\n4. Benchmark with a Deterministic Mini Suite\n\nUse the same prompt set and output checks across candidates so results are comparable.\n\nMinimum benchmark set:\n\nOne typical request\nOne edge-case request\nOne failure-prone request\n5. Minimize External Data\n\nSend only what is required for the selected endpoint.\n\nNever send credentials, local paths, or unrelated private context in request payloads.\n\n6. Use a Fallback Ladder\n\nIf the preferred model fails, apply ordered fallback:\n\nRetry same endpoint with smaller payload\nSwitch to a compatible backup model\nSwitch to local-only workflow if available\n7. Keep Runs Reproducible\n\nLog selected model id, endpoint, key parameters, and evaluation result in local memory so future runs are consistent and auditable.\n\nCommon Traps\nPicking the highest download count as the only criterion -> often misses license, latency, or domain fit.\nIgnoring gated model requirements -> integration fails at runtime due to access restrictions.\nComparing models with different prompts -> quality conclusions become unreliable.\nSending full user context to inference endpoints -> unnecessary privacy exposure.\nSkipping fallback design -> workflows fail hard on transient endpoint errors.\nExternal Endpoints\n\nUse discovery endpoints before inference so candidate selection remains explainable and reproducible.\n\nEndpoint\tData Sent\tPurpose\nhttps://huggingface.co/api/models\tSearch terms, filter parameters\tDiscover model candidates\nhttps://huggingface.co/api/datasets\tSearch terms, filter parameters\tDiscover dataset candidates\nhttps://huggingface.co/api/spaces\tSearch terms, filter parameters\tDiscover runnable Spaces\nhttps://api-inference.huggingface.co/models/{model_id}\tPrompt or task input payload, selected model id, auth token\tRun hosted inference\n\nNo other data is sent externally.\n\nSecurity & Privacy\n\nData that leaves your machine:\n\nSearch terms and filter inputs sent to Hugging Face discovery APIs.\nInference payloads sent to Hugging Face Inference API when execution is requested.\n\nData that stays local:\n\nPreferences, shortlists, evaluation notes, and endpoint decisions in ~/hugging-face/.\n\nThis skill does NOT:\n\nExfiltrate local files by default.\nSend undeclared network requests.\nStore raw secrets in local notes.\nModify its own skill definition file.\nTrust\n\nBy using this skill, selected request data is sent to Hugging Face services. Only install if you trust Hugging Face with the inputs you choose to process.\n\nRelated Skills\n\nInstall with clawhub install <slug> if user confirms:\n\nai - general AI strategy and model-selection framing\napi - API-first integration patterns and HTTP debugging\ndata-analysis - dataset inspection and quality interpretation\ndata - structured data workflows and extraction patterns\ncode - implementation support for scripts and adapters\nFeedback\nIf useful: clawhub star hugging-face\nStay updated: clawhub sync"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/ivangdavila/hugging-face",
    "publisherUrl": "https://clawhub.ai/ivangdavila/hugging-face",
    "owner": "ivangdavila",
    "version": "1.0.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/hugging-face",
    "downloadUrl": "https://openagent3.xyz/downloads/hugging-face",
    "agentUrl": "https://openagent3.xyz/skills/hugging-face/agent",
    "manifestUrl": "https://openagent3.xyz/skills/hugging-face/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/hugging-face/agent.md"
  }
}