{
  "schemaVersion": "1.0",
  "item": {
    "slug": "pref0",
    "name": "pref0",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/fliellerjulian/pref0",
    "canonicalUrl": "https://clawhub.ai/fliellerjulian/pref0",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/pref0",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=pref0",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md",
      "notes.txt"
    ],
    "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/pref0"
    },
    "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/pref0",
    "agentPageUrl": "https://openagent3.xyz/skills/pref0/agent",
    "manifestUrl": "https://openagent3.xyz/skills/pref0/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/pref0/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": "pref0 — Preference Learning for AI Agents",
        "body": "You have access to the pref0 API. It learns user preferences from conversations and serves them back at inference time. The more conversations you track, the better it gets."
      },
      {
        "title": "After a conversation ends → Track it",
        "body": "After finishing a conversation (or at natural breakpoints), send the messages to pref0 so it can extract preferences. This is especially valuable when the user corrects you (e.g., \"use pnpm, not npm\") or states explicit preferences (e.g., \"always use metric units\")."
      },
      {
        "title": "Before responding to a user → Fetch their preferences",
        "body": "Before generating a response, fetch the user's learned preferences and follow them. This prevents the user from having to repeat themselves across sessions."
      },
      {
        "title": "API Reference",
        "body": "Base URL: https://api.pref0.com\nAuth: Authorization: Bearer $PREF0_API_KEY"
      },
      {
        "title": "Track a conversation (POST /v1/track)",
        "body": "Send a conversation so pref0 can learn from it. It extracts corrections, explicit preferences, and behavioral patterns automatically.\n\ncurl -X POST https://api.pref0.com/v1/track \\\n  -H \"Authorization: Bearer $PREF0_API_KEY\" \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\n    \"userId\": \"<user-id>\",\n    \"messages\": [\n      { \"role\": \"user\", \"content\": \"Help me set up a new project\" },\n      { \"role\": \"assistant\", \"content\": \"Here is a project using npm and JavaScript...\" },\n      { \"role\": \"user\", \"content\": \"Use pnpm, not npm. And TypeScript.\" },\n      { \"role\": \"assistant\", \"content\": \"Updated to pnpm and TypeScript...\" }\n    ]\n  }'\n\nResponse:\n\n{\n  \"messagesAnalyzed\": 4,\n  \"preferences\": { \"created\": 2, \"reinforced\": 0, \"decreased\": 0, \"removed\": 0 },\n  \"patterns\": { \"created\": 1, \"reinforced\": 0 }\n}\n\nThe response tells you how many messages were processed (messagesAnalyzed) and exactly what changed: created (new preference learned), reinforced (existing preference seen again, confidence increased), decreased (user retracted, confidence lowered), removed (fully retracted and deleted)."
      },
      {
        "title": "Get learned preferences (GET /v1/profiles/:userId)",
        "body": "Retrieve the user's learned preference profile. Use ?minConfidence=0.5 to only get well-learned preferences suitable for system prompt injection.\n\ncurl https://api.pref0.com/v1/profiles/<user-id>?minConfidence=0.5 \\\n  -H \"Authorization: Bearer $PREF0_API_KEY\"\n\nResponse:\n\n{\n  \"userId\": \"user_abc123\",\n  \"preferences\": [\n    {\n      \"key\": \"language\",\n      \"value\": \"typescript\",\n      \"confidence\": 0.85,\n      \"evidence\": \"User said: Use TypeScript, not JavaScript\",\n      \"firstSeen\": \"2026-01-15T10:00:00.000Z\",\n      \"lastSeen\": \"2026-02-05T14:30:00.000Z\"\n    },\n    {\n      \"key\": \"package_manager\",\n      \"value\": \"pnpm\",\n      \"confidence\": 0.85,\n      \"evidence\": \"User said: use pnpm instead of npm\",\n      \"firstSeen\": \"2026-01-15T10:00:00.000Z\",\n      \"lastSeen\": \"2026-02-03T09:15:00.000Z\"\n    },\n    {\n      \"key\": \"css_framework\",\n      \"value\": \"tailwind\",\n      \"confidence\": 0.70,\n      \"evidence\": \"User said: Use Tailwind, not Bootstrap\",\n      \"firstSeen\": \"2026-01-20T16:45:00.000Z\",\n      \"lastSeen\": \"2026-01-20T16:45:00.000Z\"\n    }\n  ],\n  \"patterns\": [\n    { \"pattern\": \"prefers explicit tooling choices over defaults\", \"confidence\": 0.60 }\n  ],\n  \"prompt\": \"The following preferences have been learned from this user's previous conversations. Follow them unless explicitly told otherwise:\\n- language: typescript\\n- package_manager: pnpm\\n- css_framework: tailwind\\n\\nBehavioral patterns observed:\\n- prefers explicit tooling choices over defaults\"\n}\n\nEach preference includes evidence (the quote that triggered extraction), firstSeen (when first learned), and lastSeen (when last reinforced). The prompt field is a ready-to-use string you can append directly to your system prompt."
      },
      {
        "title": "Delete a user profile (DELETE /v1/profiles/:userId)",
        "body": "Reset a user's learned preferences. Use for preference resets or data deletion requests.\n\ncurl -X DELETE https://api.pref0.com/v1/profiles/<user-id> \\\n  -H \"Authorization: Bearer $PREF0_API_KEY\"\n\nReturns 204 No Content."
      },
      {
        "title": "How to integrate into your workflow",
        "body": "Identify the user. Use a stable user ID (email, account ID, phone number — whatever you have).\n\n\nAt the start of a session, fetch preferences:\n\nCall GET /v1/profiles/{userId}?minConfidence=0.5\nUse the prompt field to inject into your system prompt directly, or use the structured preferences array for more control.\n\n\n\nAt the end of a session, track the conversation:\n\nCall POST /v1/track with the full message history\npref0 handles extraction and confidence scoring automatically\n\n\n\nPreferences compound over time. Corrections start at 0.70 confidence, implied preferences at 0.40. Each repeated signal adds +0.15, capped at 1.0."
      },
      {
        "title": "Confidence guide",
        "body": "Signal typeStarting confidenceExampleExplicit correction0.70\"Use Tailwind, not Bootstrap\"Implied preference0.40\"Deploy it to Vercel\"Behavioral pattern0.30User consistently wants short repliesEach repeat+0.15Same preference across sessions"
      },
      {
        "title": "Setup",
        "body": "Sign up at pref0.com\nCreate an API key in the dashboard\nSet the PREF0_API_KEY environment variable\nFirst 100 requests/month are free, then $5 per 1,000 requests"
      }
    ],
    "body": "pref0 — Preference Learning for AI Agents\n\nYou have access to the pref0 API. It learns user preferences from conversations and serves them back at inference time. The more conversations you track, the better it gets.\n\nWhen to use this skill\nAfter a conversation ends → Track it\n\nAfter finishing a conversation (or at natural breakpoints), send the messages to pref0 so it can extract preferences. This is especially valuable when the user corrects you (e.g., \"use pnpm, not npm\") or states explicit preferences (e.g., \"always use metric units\").\n\nBefore responding to a user → Fetch their preferences\n\nBefore generating a response, fetch the user's learned preferences and follow them. This prevents the user from having to repeat themselves across sessions.\n\nAPI Reference\n\nBase URL: https://api.pref0.com Auth: Authorization: Bearer $PREF0_API_KEY\n\nTrack a conversation (POST /v1/track)\n\nSend a conversation so pref0 can learn from it. It extracts corrections, explicit preferences, and behavioral patterns automatically.\n\ncurl -X POST https://api.pref0.com/v1/track \\\n  -H \"Authorization: Bearer $PREF0_API_KEY\" \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\n    \"userId\": \"<user-id>\",\n    \"messages\": [\n      { \"role\": \"user\", \"content\": \"Help me set up a new project\" },\n      { \"role\": \"assistant\", \"content\": \"Here is a project using npm and JavaScript...\" },\n      { \"role\": \"user\", \"content\": \"Use pnpm, not npm. And TypeScript.\" },\n      { \"role\": \"assistant\", \"content\": \"Updated to pnpm and TypeScript...\" }\n    ]\n  }'\n\n\nResponse:\n\n{\n  \"messagesAnalyzed\": 4,\n  \"preferences\": { \"created\": 2, \"reinforced\": 0, \"decreased\": 0, \"removed\": 0 },\n  \"patterns\": { \"created\": 1, \"reinforced\": 0 }\n}\n\n\nThe response tells you how many messages were processed (messagesAnalyzed) and exactly what changed: created (new preference learned), reinforced (existing preference seen again, confidence increased), decreased (user retracted, confidence lowered), removed (fully retracted and deleted).\n\nGet learned preferences (GET /v1/profiles/:userId)\n\nRetrieve the user's learned preference profile. Use ?minConfidence=0.5 to only get well-learned preferences suitable for system prompt injection.\n\ncurl https://api.pref0.com/v1/profiles/<user-id>?minConfidence=0.5 \\\n  -H \"Authorization: Bearer $PREF0_API_KEY\"\n\n\nResponse:\n\n{\n  \"userId\": \"user_abc123\",\n  \"preferences\": [\n    {\n      \"key\": \"language\",\n      \"value\": \"typescript\",\n      \"confidence\": 0.85,\n      \"evidence\": \"User said: Use TypeScript, not JavaScript\",\n      \"firstSeen\": \"2026-01-15T10:00:00.000Z\",\n      \"lastSeen\": \"2026-02-05T14:30:00.000Z\"\n    },\n    {\n      \"key\": \"package_manager\",\n      \"value\": \"pnpm\",\n      \"confidence\": 0.85,\n      \"evidence\": \"User said: use pnpm instead of npm\",\n      \"firstSeen\": \"2026-01-15T10:00:00.000Z\",\n      \"lastSeen\": \"2026-02-03T09:15:00.000Z\"\n    },\n    {\n      \"key\": \"css_framework\",\n      \"value\": \"tailwind\",\n      \"confidence\": 0.70,\n      \"evidence\": \"User said: Use Tailwind, not Bootstrap\",\n      \"firstSeen\": \"2026-01-20T16:45:00.000Z\",\n      \"lastSeen\": \"2026-01-20T16:45:00.000Z\"\n    }\n  ],\n  \"patterns\": [\n    { \"pattern\": \"prefers explicit tooling choices over defaults\", \"confidence\": 0.60 }\n  ],\n  \"prompt\": \"The following preferences have been learned from this user's previous conversations. Follow them unless explicitly told otherwise:\\n- language: typescript\\n- package_manager: pnpm\\n- css_framework: tailwind\\n\\nBehavioral patterns observed:\\n- prefers explicit tooling choices over defaults\"\n}\n\n\nEach preference includes evidence (the quote that triggered extraction), firstSeen (when first learned), and lastSeen (when last reinforced). The prompt field is a ready-to-use string you can append directly to your system prompt.\n\nDelete a user profile (DELETE /v1/profiles/:userId)\n\nReset a user's learned preferences. Use for preference resets or data deletion requests.\n\ncurl -X DELETE https://api.pref0.com/v1/profiles/<user-id> \\\n  -H \"Authorization: Bearer $PREF0_API_KEY\"\n\n\nReturns 204 No Content.\n\nHow to integrate into your workflow\n\nIdentify the user. Use a stable user ID (email, account ID, phone number — whatever you have).\n\nAt the start of a session, fetch preferences:\n\nCall GET /v1/profiles/{userId}?minConfidence=0.5\nUse the prompt field to inject into your system prompt directly, or use the structured preferences array for more control.\n\nAt the end of a session, track the conversation:\n\nCall POST /v1/track with the full message history\npref0 handles extraction and confidence scoring automatically\n\nPreferences compound over time. Corrections start at 0.70 confidence, implied preferences at 0.40. Each repeated signal adds +0.15, capped at 1.0.\n\nConfidence guide\nSignal type\tStarting confidence\tExample\nExplicit correction\t0.70\t\"Use Tailwind, not Bootstrap\"\nImplied preference\t0.40\t\"Deploy it to Vercel\"\nBehavioral pattern\t0.30\tUser consistently wants short replies\nEach repeat\t+0.15\tSame preference across sessions\nSetup\nSign up at pref0.com\nCreate an API key in the dashboard\nSet the PREF0_API_KEY environment variable\nFirst 100 requests/month are free, then $5 per 1,000 requests"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/fliellerjulian/pref0",
    "publisherUrl": "https://clawhub.ai/fliellerjulian/pref0",
    "owner": "fliellerjulian",
    "version": "1.0.1",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/pref0",
    "downloadUrl": "https://openagent3.xyz/downloads/pref0",
    "agentUrl": "https://openagent3.xyz/skills/pref0/agent",
    "manifestUrl": "https://openagent3.xyz/skills/pref0/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/pref0/agent.md"
  }
}