{
  "schemaVersion": "1.0",
  "item": {
    "slug": "oracle",
    "name": "Oracle",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/steipete/oracle",
    "canonicalUrl": "https://clawhub.ai/steipete/oracle",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/oracle",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=oracle",
    "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/oracle"
    },
    "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/oracle",
    "agentPageUrl": "https://openagent3.xyz/skills/oracle/agent",
    "manifestUrl": "https://openagent3.xyz/skills/oracle/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/oracle/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": "Oracle (CLI) — best use",
        "body": "Oracle bundles your prompt + selected files into one “one-shot” request so another model can answer with real repo context (API or browser automation). Treat outputs as advisory: verify against the codebase + tests."
      },
      {
        "title": "Main use case (browser, GPT‑5.2 Pro)",
        "body": "Default workflow here: --engine browser with GPT‑5.2 Pro in ChatGPT. This is the “human in the loop” path: it can take ~10 minutes to ~1 hour; expect a stored session you can reattach to.\n\nRecommended defaults:\n\nEngine: browser (--engine browser)\nModel: GPT‑5.2 Pro (either --model gpt-5.2-pro or a ChatGPT picker label like --model \"5.2 Pro\")\nAttachments: directories/globs + excludes; avoid secrets."
      },
      {
        "title": "Golden path (fast + reliable)",
        "body": "Pick a tight file set (fewest files that still contain the truth).\nPreview what you’re about to send (--dry-run + --files-report when needed).\nRun in browser mode for the usual GPT‑5.2 Pro ChatGPT workflow; use API only when you explicitly want it.\nIf the run detaches/timeouts: reattach to the stored session (don’t re-run)."
      },
      {
        "title": "Commands (preferred)",
        "body": "Show help (once/session):\n\nnpx -y @steipete/oracle --help\n\n\n\nPreview (no tokens):\n\nnpx -y @steipete/oracle --dry-run summary -p \"<task>\" --file \"src/**\" --file \"!**/*.test.*\"\nnpx -y @steipete/oracle --dry-run full -p \"<task>\" --file \"src/**\"\n\n\n\nToken/cost sanity:\n\nnpx -y @steipete/oracle --dry-run summary --files-report -p \"<task>\" --file \"src/**\"\n\n\n\nBrowser run (main path; long-running is normal):\n\nnpx -y @steipete/oracle --engine browser --model gpt-5.2-pro -p \"<task>\" --file \"src/**\"\n\n\n\nManual paste fallback (assemble bundle, copy to clipboard):\n\nnpx -y @steipete/oracle --render --copy -p \"<task>\" --file \"src/**\"\nNote: --copy is a hidden alias for --copy-markdown."
      },
      {
        "title": "Attaching files (--file)",
        "body": "--file accepts files, directories, and globs. You can pass it multiple times; entries can be comma-separated.\n\nInclude:\n\n--file \"src/**\" (directory glob)\n--file src/index.ts (literal file)\n--file docs --file README.md (literal directory + file)\n\n\n\nExclude (prefix with !):\n\n--file \"src/**\" --file \"!src/**/*.test.ts\" --file \"!**/*.snap\"\n\n\n\nDefaults (important behavior from the implementation):\n\nDefault-ignored dirs: node_modules, dist, coverage, .git, .turbo, .next, build, tmp (skipped unless you explicitly pass them as literal dirs/files).\nHonors .gitignore when expanding globs.\nDoes not follow symlinks (glob expansion uses followSymbolicLinks: false).\nDotfiles are filtered unless you explicitly opt in with a pattern that includes a dot-segment (e.g. --file \".github/**\").\nHard cap: files > 1 MB are rejected (split files or narrow the match)."
      },
      {
        "title": "Budget + observability",
        "body": "Target: keep total input under ~196k tokens.\nUse --files-report (and/or --dry-run json) to spot the token hogs before spending.\nIf you need hidden/advanced knobs: npx -y @steipete/oracle --help --verbose."
      },
      {
        "title": "Engines (API vs browser)",
        "body": "Auto-pick: uses api when OPENAI_API_KEY is set, otherwise browser.\nBrowser engine supports GPT + Gemini only; use --engine api for Claude/Grok/Codex or multi-model runs.\nAPI runs require explicit user consent before starting because they incur usage costs.\nBrowser attachments:\n\n--browser-attachments auto|never|always (auto pastes inline up to ~60k chars then uploads).\n\n\nRemote browser host (signed-in machine runs automation):\n\nHost: oracle serve --host 0.0.0.0 --port 9473 --token <secret>\nClient: oracle --engine browser --remote-host <host:port> --remote-token <secret> -p \"<task>\" --file \"src/**\""
      },
      {
        "title": "Sessions + slugs (don’t lose work)",
        "body": "Stored under ~/.oracle/sessions (override with ORACLE_HOME_DIR).\nRuns may detach or take a long time (browser + GPT‑5.2 Pro often does). If the CLI times out: don’t re-run; reattach.\n\nList: oracle status --hours 72\nAttach: oracle session <id> --render\n\n\nUse --slug \"<3-5 words>\" to keep session IDs readable.\nDuplicate prompt guard exists; use --force only when you truly want a fresh run."
      },
      {
        "title": "Prompt template (high signal)",
        "body": "Oracle starts with zero project knowledge. Assume the model cannot infer your stack, build tooling, conventions, or “obvious” paths. Include:\n\nProject briefing (stack + build/test commands + platform constraints).\n“Where things live” (key directories, entrypoints, config files, dependency boundaries).\nExact question + what you tried + the error text (verbatim).\nConstraints (“don’t change X”, “must keep public API”, “perf budget”, etc).\nDesired output (“return patch plan + tests”, “list risky assumptions”, “give 3 options with tradeoffs”)."
      },
      {
        "title": "“Exhaustive prompt” pattern (for later restoration)",
        "body": "When you know this will be a long investigation, write a prompt that can stand alone later:\n\nTop: 6–30 sentence project briefing + current goal.\nMiddle: concrete repro steps + exact errors + what you already tried.\nBottom: attach all context files needed so a fresh model can fully understand (entrypoints, configs, key modules, docs).\n\nIf you need to reproduce the same context later, re-run with the same prompt + --file … set (Oracle runs are one-shot; the model doesn’t remember prior runs)."
      },
      {
        "title": "Safety",
        "body": "Don’t attach secrets by default (.env, key files, auth tokens). Redact aggressively; share only what’s required.\nPrefer “just enough context”: fewer files + better prompt beats whole-repo dumps."
      }
    ],
    "body": "Oracle (CLI) — best use\n\nOracle bundles your prompt + selected files into one “one-shot” request so another model can answer with real repo context (API or browser automation). Treat outputs as advisory: verify against the codebase + tests.\n\nMain use case (browser, GPT‑5.2 Pro)\n\nDefault workflow here: --engine browser with GPT‑5.2 Pro in ChatGPT. This is the “human in the loop” path: it can take ~10 minutes to ~1 hour; expect a stored session you can reattach to.\n\nRecommended defaults:\n\nEngine: browser (--engine browser)\nModel: GPT‑5.2 Pro (either --model gpt-5.2-pro or a ChatGPT picker label like --model \"5.2 Pro\")\nAttachments: directories/globs + excludes; avoid secrets.\nGolden path (fast + reliable)\nPick a tight file set (fewest files that still contain the truth).\nPreview what you’re about to send (--dry-run + --files-report when needed).\nRun in browser mode for the usual GPT‑5.2 Pro ChatGPT workflow; use API only when you explicitly want it.\nIf the run detaches/timeouts: reattach to the stored session (don’t re-run).\nCommands (preferred)\n\nShow help (once/session):\n\nnpx -y @steipete/oracle --help\n\nPreview (no tokens):\n\nnpx -y @steipete/oracle --dry-run summary -p \"<task>\" --file \"src/**\" --file \"!**/*.test.*\"\nnpx -y @steipete/oracle --dry-run full -p \"<task>\" --file \"src/**\"\n\nToken/cost sanity:\n\nnpx -y @steipete/oracle --dry-run summary --files-report -p \"<task>\" --file \"src/**\"\n\nBrowser run (main path; long-running is normal):\n\nnpx -y @steipete/oracle --engine browser --model gpt-5.2-pro -p \"<task>\" --file \"src/**\"\n\nManual paste fallback (assemble bundle, copy to clipboard):\n\nnpx -y @steipete/oracle --render --copy -p \"<task>\" --file \"src/**\"\nNote: --copy is a hidden alias for --copy-markdown.\nAttaching files (--file)\n\n--file accepts files, directories, and globs. You can pass it multiple times; entries can be comma-separated.\n\nInclude:\n\n--file \"src/**\" (directory glob)\n--file src/index.ts (literal file)\n--file docs --file README.md (literal directory + file)\n\nExclude (prefix with !):\n\n--file \"src/**\" --file \"!src/**/*.test.ts\" --file \"!**/*.snap\"\n\nDefaults (important behavior from the implementation):\n\nDefault-ignored dirs: node_modules, dist, coverage, .git, .turbo, .next, build, tmp (skipped unless you explicitly pass them as literal dirs/files).\nHonors .gitignore when expanding globs.\nDoes not follow symlinks (glob expansion uses followSymbolicLinks: false).\nDotfiles are filtered unless you explicitly opt in with a pattern that includes a dot-segment (e.g. --file \".github/**\").\nHard cap: files > 1 MB are rejected (split files or narrow the match).\nBudget + observability\nTarget: keep total input under ~196k tokens.\nUse --files-report (and/or --dry-run json) to spot the token hogs before spending.\nIf you need hidden/advanced knobs: npx -y @steipete/oracle --help --verbose.\nEngines (API vs browser)\nAuto-pick: uses api when OPENAI_API_KEY is set, otherwise browser.\nBrowser engine supports GPT + Gemini only; use --engine api for Claude/Grok/Codex or multi-model runs.\nAPI runs require explicit user consent before starting because they incur usage costs.\nBrowser attachments:\n--browser-attachments auto|never|always (auto pastes inline up to ~60k chars then uploads).\nRemote browser host (signed-in machine runs automation):\nHost: oracle serve --host 0.0.0.0 --port 9473 --token <secret>\nClient: oracle --engine browser --remote-host <host:port> --remote-token <secret> -p \"<task>\" --file \"src/**\"\nSessions + slugs (don’t lose work)\nStored under ~/.oracle/sessions (override with ORACLE_HOME_DIR).\nRuns may detach or take a long time (browser + GPT‑5.2 Pro often does). If the CLI times out: don’t re-run; reattach.\nList: oracle status --hours 72\nAttach: oracle session <id> --render\nUse --slug \"<3-5 words>\" to keep session IDs readable.\nDuplicate prompt guard exists; use --force only when you truly want a fresh run.\nPrompt template (high signal)\n\nOracle starts with zero project knowledge. Assume the model cannot infer your stack, build tooling, conventions, or “obvious” paths. Include:\n\nProject briefing (stack + build/test commands + platform constraints).\n“Where things live” (key directories, entrypoints, config files, dependency boundaries).\nExact question + what you tried + the error text (verbatim).\nConstraints (“don’t change X”, “must keep public API”, “perf budget”, etc).\nDesired output (“return patch plan + tests”, “list risky assumptions”, “give 3 options with tradeoffs”).\n“Exhaustive prompt” pattern (for later restoration)\n\nWhen you know this will be a long investigation, write a prompt that can stand alone later:\n\nTop: 6–30 sentence project briefing + current goal.\nMiddle: concrete repro steps + exact errors + what you already tried.\nBottom: attach all context files needed so a fresh model can fully understand (entrypoints, configs, key modules, docs).\n\nIf you need to reproduce the same context later, re-run with the same prompt + --file … set (Oracle runs are one-shot; the model doesn’t remember prior runs).\n\nSafety\nDon’t attach secrets by default (.env, key files, auth tokens). Redact aggressively; share only what’s required.\nPrefer “just enough context”: fewer files + better prompt beats whole-repo dumps."
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/steipete/oracle",
    "publisherUrl": "https://clawhub.ai/steipete/oracle",
    "owner": "steipete",
    "version": "1.0.1",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/oracle",
    "downloadUrl": "https://openagent3.xyz/downloads/oracle",
    "agentUrl": "https://openagent3.xyz/skills/oracle/agent",
    "manifestUrl": "https://openagent3.xyz/skills/oracle/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/oracle/agent.md"
  }
}