{
  "schemaVersion": "1.0",
  "item": {
    "slug": "smart-model-switching",
    "name": "Smart Model Switching",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/millibus/smart-model-switching",
    "canonicalUrl": "https://clawhub.ai/millibus/smart-model-switching",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/smart-model-switching",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=smart-model-switching",
    "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": "2026-05-07T17:22:31.273Z",
      "expiresAt": "2026-05-14T17:22:31.273Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=afrexai-annual-report",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=afrexai-annual-report",
        "contentDisposition": "attachment; filename=\"afrexai-annual-report-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/smart-model-switching"
    },
    "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/smart-model-switching",
    "agentPageUrl": "https://openagent3.xyz/skills/smart-model-switching/agent",
    "manifestUrl": "https://openagent3.xyz/skills/smart-model-switching/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/smart-model-switching/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": "Smart Model Switching",
        "body": "Three-tier Claude routing: Haiku → Sonnet → Opus\n\nStart with the cheapest model. Escalate only when needed. Save 50-90% on API costs."
      },
      {
        "title": "The Golden Rule",
        "body": "If a human would need more than 30 seconds of focused thinking, escalate from Haiku to Sonnet.\nIf the task involves architecture, complex tradeoffs, or deep reasoning, escalate to Opus."
      },
      {
        "title": "Cost Reality",
        "body": "ModelInputOutputRelative CostHaiku$0.25/M$1.25/M1x (baseline)Sonnet$3.00/M$15.00/M12xOpus$15.00/M$75.00/M60x\n\nBottom line: Wrong model selection wastes money OR time. Haiku for simple, Sonnet for standard, Opus for complex."
      },
      {
        "title": "💚 HAIKU — Default for Simple Tasks",
        "body": "Stay on Haiku for:\n\nFactual Q&A — \"what is X\", \"who is Y\", \"when did Z\"\nQuick lookups — definitions, unit conversions, short translations\nStatus checks — calendar, file reads, session monitoring\nHeartbeats — periodic checks, HEARTBEAT_OK responses\nMemory & reminders — \"remember this\", \"remind me to...\"\nCasual conversation — greetings, small talk, acknowledgments\nSimple file ops — read, list, basic writes\nOne-liner tasks — anything answerable in 1-2 sentences"
      },
      {
        "title": "NEVER do these on Haiku",
        "body": "❌ Write code longer than 10 lines\n❌ Create comparison tables\n❌ Write more than 3 paragraphs\n❌ Do multi-step analysis\n❌ Write reports or proposals"
      },
      {
        "title": "💛 SONNET — Standard Work (The Workhorse)",
        "body": "Escalate to Sonnet for:"
      },
      {
        "title": "Code & Technical",
        "body": "Code generation — write functions, build features, scripts\nCode review — PR reviews, quality checks\nDebugging — standard bug investigation\nDocumentation — README, comments, user guides"
      },
      {
        "title": "Analysis & Planning",
        "body": "Analysis & evaluation — compare options, assess trade-offs\nPlanning — project plans, roadmaps, task breakdowns\nResearch synthesis — combining multiple sources\nMulti-step reasoning — \"first... then... finally\""
      },
      {
        "title": "Writing & Content",
        "body": "Long-form writing — reports, proposals, articles (>3 paragraphs)\nCreative writing — blog posts, descriptions, copy\nSummarization — long documents, transcripts\nStructured output — tables, outlines, formatted docs"
      },
      {
        "title": "❤️ OPUS — Complex Reasoning Only",
        "body": "Escalate to Opus for:"
      },
      {
        "title": "Architecture & Design",
        "body": "System architecture decisions\nMajor codebase refactoring\nDesign pattern selection with tradeoffs\nDatabase schema design"
      },
      {
        "title": "Deep Analysis",
        "body": "Complex debugging (multi-file, race conditions)\nSecurity reviews\nPerformance optimization strategy\nRoot cause analysis of subtle bugs"
      },
      {
        "title": "Strategic & Creative",
        "body": "Strategic planning — business decisions, roadmaps\nNuanced judgment — ethics, ambiguity, competing values\nDeep research — comprehensive multi-source analysis"
      },
      {
        "title": "For Subagents",
        "body": "```javascript\n// Routine monitoring\nsessions_spawn(task=\"Check backup status\", model=\"haiku\")\n\n// Standard code work\nsessions_spawn(task=\"Build the REST API endpoint\", model=\"sonnet\")\n\n// Architecture decisions\nsessions_spawn(task=\"Design the database schema for multi-tenancy\", model=\"opus\")\n```"
      },
      {
        "title": "For Cron Jobs",
        "body": "```json\n{\n\"payload\": {\n\"kind\": \"agentTurn\",\n\"model\": \"haiku\"\n}\n}\n```\nAlways use Haiku for cron unless the task genuinely needs reasoning."
      },
      {
        "title": "📊 Quick Decision Tree",
        "body": "```\nIs it a greeting, lookup, status check, or 1-2 sentence answer?\nYES → HAIKU\nNO ↓\n\nIs it code, analysis, planning, writing, or multi-step?\nYES → SONNET\nNO ↓\n\nIs it architecture, deep reasoning, or critical decision?\nYES → OPUS\nNO → Default to SONNET, escalate if struggling\n```"
      },
      {
        "title": "📋 Quick Reference Card",
        "body": "```\n┌─────────────────────────────────────────────────────────────┐\n│                  SMART MODEL SWITCHING                      │\n│                  Haiku → Sonnet → Opus                      │\n├─────────────────────────────────────────────────────────────┤\n│  💚 HAIKU (cheapest)                                        │\n│  • Greetings, status checks, quick lookups                  │\n│  • Factual Q&A, definitions, reminders                      │\n│  • Simple file ops, 1-2 sentence answers                    │\n├─────────────────────────────────────────────────────────────┤\n│  💛 SONNET (standard)                                       │\n│  • Code > 10 lines, debugging                               │\n│  • Analysis, comparisons, planning                          │\n│  • Reports, proposals, long writing                         │\n├─────────────────────────────────────────────────────────────┤\n│  ❤️ OPUS (complex)                                          │\n│  • Architecture decisions                                   │\n│  • Complex debugging, multi-file refactoring                │\n│  • Strategic planning, deep research                        │\n├─────────────────────────────────────────────────────────────┤\n│  💡 RULE: If a human needs > 30 sec thinking → escalate     │\n│  💰 COST: Haiku 1x → Sonnet 12x → Opus 60x                  │\n└─────────────────────────────────────────────────────────────┘\n```\n\nBuilt for Claude-only setups with Haiku, Sonnet, and Opus.\nInspired by save-money skill, extended with three-tier progression."
      }
    ],
    "body": "Smart Model Switching\n\nThree-tier Claude routing: Haiku → Sonnet → Opus\n\nStart with the cheapest model. Escalate only when needed. Save 50-90% on API costs.\n\nThe Golden Rule\n\nIf a human would need more than 30 seconds of focused thinking, escalate from Haiku to Sonnet. If the task involves architecture, complex tradeoffs, or deep reasoning, escalate to Opus.\n\nCost Reality\nModel\tInput\tOutput\tRelative Cost\nHaiku\t$0.25/M\t$1.25/M\t1x (baseline)\nSonnet\t$3.00/M\t$15.00/M\t12x\nOpus\t$15.00/M\t$75.00/M\t60x\n\nBottom line: Wrong model selection wastes money OR time. Haiku for simple, Sonnet for standard, Opus for complex.\n\n💚 HAIKU — Default for Simple Tasks\n\nStay on Haiku for:\n\nFactual Q&A — \"what is X\", \"who is Y\", \"when did Z\"\nQuick lookups — definitions, unit conversions, short translations\nStatus checks — calendar, file reads, session monitoring\nHeartbeats — periodic checks, HEARTBEAT_OK responses\nMemory & reminders — \"remember this\", \"remind me to...\"\nCasual conversation — greetings, small talk, acknowledgments\nSimple file ops — read, list, basic writes\nOne-liner tasks — anything answerable in 1-2 sentences\nNEVER do these on Haiku\n❌ Write code longer than 10 lines\n❌ Create comparison tables\n❌ Write more than 3 paragraphs\n❌ Do multi-step analysis\n❌ Write reports or proposals\n💛 SONNET — Standard Work (The Workhorse)\n\nEscalate to Sonnet for:\n\nCode & Technical\nCode generation — write functions, build features, scripts\nCode review — PR reviews, quality checks\nDebugging — standard bug investigation\nDocumentation — README, comments, user guides\nAnalysis & Planning\nAnalysis & evaluation — compare options, assess trade-offs\nPlanning — project plans, roadmaps, task breakdowns\nResearch synthesis — combining multiple sources\nMulti-step reasoning — \"first... then... finally\"\nWriting & Content\nLong-form writing — reports, proposals, articles (>3 paragraphs)\nCreative writing — blog posts, descriptions, copy\nSummarization — long documents, transcripts\nStructured output — tables, outlines, formatted docs\n❤️ OPUS — Complex Reasoning Only\n\nEscalate to Opus for:\n\nArchitecture & Design\nSystem architecture decisions\nMajor codebase refactoring\nDesign pattern selection with tradeoffs\nDatabase schema design\nDeep Analysis\nComplex debugging (multi-file, race conditions)\nSecurity reviews\nPerformance optimization strategy\nRoot cause analysis of subtle bugs\nStrategic & Creative\nStrategic planning — business decisions, roadmaps\nNuanced judgment — ethics, ambiguity, competing values\nDeep research — comprehensive multi-source analysis\n🔄 Implementation\nFor Subagents\n\n```javascript // Routine monitoring sessions_spawn(task=\"Check backup status\", model=\"haiku\")\n\n// Standard code work\nsessions_spawn(task=\"Build the REST API endpoint\", model=\"sonnet\")\n\n// Architecture decisions sessions_spawn(task=\"Design the database schema for multi-tenancy\", model=\"opus\") ```\n\nFor Cron Jobs\n\n```json { \"payload\": { \"kind\": \"agentTurn\", \"model\": \"haiku\" } } ``` Always use Haiku for cron unless the task genuinely needs reasoning.\n\n📊 Quick Decision Tree\n\n``` Is it a greeting, lookup, status check, or 1-2 sentence answer? YES → HAIKU NO ↓\n\nIs it code, analysis, planning, writing, or multi-step? YES → SONNET\nNO ↓\n\nIs it architecture, deep reasoning, or critical decision? YES → OPUS NO → Default to SONNET, escalate if struggling ```\n\n📋 Quick Reference Card\n\n``` ┌─────────────────────────────────────────────────────────────┐ │ SMART MODEL SWITCHING │ │ Haiku → Sonnet → Opus │ ├─────────────────────────────────────────────────────────────┤ │ 💚 HAIKU (cheapest) │ │ • Greetings, status checks, quick lookups │ │ • Factual Q&A, definitions, reminders │ │ • Simple file ops, 1-2 sentence answers │ ├─────────────────────────────────────────────────────────────┤ │ 💛 SONNET (standard) │ │ • Code > 10 lines, debugging │ │ • Analysis, comparisons, planning │ │ • Reports, proposals, long writing │ ├─────────────────────────────────────────────────────────────┤ │ ❤️ OPUS (complex) │ │ • Architecture decisions │ │ • Complex debugging, multi-file refactoring │ │ • Strategic planning, deep research │ ├─────────────────────────────────────────────────────────────┤ │ 💡 RULE: If a human needs > 30 sec thinking → escalate │ │ 💰 COST: Haiku 1x → Sonnet 12x → Opus 60x │ └─────────────────────────────────────────────────────────────┘ ```\n\nBuilt for Claude-only setups with Haiku, Sonnet, and Opus. Inspired by save-money skill, extended with three-tier progression."
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/millibus/smart-model-switching",
    "publisherUrl": "https://clawhub.ai/millibus/smart-model-switching",
    "owner": "millibus",
    "version": "1.0.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/smart-model-switching",
    "downloadUrl": "https://openagent3.xyz/downloads/smart-model-switching",
    "agentUrl": "https://openagent3.xyz/skills/smart-model-switching/agent",
    "manifestUrl": "https://openagent3.xyz/skills/smart-model-switching/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/smart-model-switching/agent.md"
  }
}