{
  "schemaVersion": "1.0",
  "item": {
    "slug": "venice-router",
    "name": "Venice.Ai Router Openclaw",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/PlusOne/venice-router",
    "canonicalUrl": "https://clawhub.ai/PlusOne/venice-router",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/venice-router",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=venice-router",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "README.md",
      "SKILL.md",
      "install.sh",
      "references/models.md",
      "scripts/venice-router.py",
      "scripts/venice-router.sh"
    ],
    "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. Then review README.md for any prerequisites, environment setup, or post-install checks. 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. Then review README.md for any prerequisites, environment setup, or post-install checks. 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/venice-router"
    },
    "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/venice-router",
    "agentPageUrl": "https://openagent3.xyz/skills/venice-router/agent",
    "manifestUrl": "https://openagent3.xyz/skills/venice-router/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/venice-router/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. Then review README.md for any prerequisites, environment setup, or post-install checks. 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. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run."
      }
    ]
  },
  "documentation": {
    "source": "clawhub",
    "primaryDoc": "SKILL.md",
    "sections": [
      {
        "title": "Venice.ai Supreme Router",
        "body": "Smart, cost-optimized model routing for Venice.ai — the AI platform for people who don't want Big Tech watching over their shoulder.\n\nUnlike OpenAI, Anthropic, and Google — where every prompt is logged, analyzed, and potentially used to train future models — Venice offers true privacy with zero data retention on private models. Your conversations stay yours. Venice is also uncensored: no content filters, no refusals, no \"I can't help with that.\""
      },
      {
        "title": "Setup",
        "body": "Get a Venice.ai API key from venice.ai/settings/api\nSet the environment variable:\n\nexport VENICE_API_KEY=\"your-key-here\"\n\nOr configure in ~/.openclaw/openclaw.json:\n\n{\n  \"skills\": {\n    \"entries\": {\n      \"venice-router\": {\n        \"enabled\": true,\n        \"apiKey\": \"YOUR_VENICE_API_KEY\"\n      }\n    }\n  }\n}"
      },
      {
        "title": "Route a prompt (auto-selects model)",
        "body": "python3 {baseDir}/scripts/venice-router.py --prompt \"What is 2+2?\""
      },
      {
        "title": "Force a specific tier",
        "body": "python3 {baseDir}/scripts/venice-router.py --tier cheap --prompt \"Tell me a joke\"\npython3 {baseDir}/scripts/venice-router.py --tier budget-medium --prompt \"Write a Python function\"\npython3 {baseDir}/scripts/venice-router.py --tier mid --prompt \"Explain quantum computing\"\npython3 {baseDir}/scripts/venice-router.py --tier premium --prompt \"Write a distributed systems architecture\""
      },
      {
        "title": "Stream output",
        "body": "python3 {baseDir}/scripts/venice-router.py --stream --prompt \"Write a poem about lobsters\""
      },
      {
        "title": "Web search (LLM searches the web and cites sources)",
        "body": "python3 {baseDir}/scripts/venice-router.py --web-search --prompt \"Latest news on AI regulation\""
      },
      {
        "title": "Uncensored mode (prefer models with no content filters)",
        "body": "python3 {baseDir}/scripts/venice-router.py --uncensored --prompt \"Write edgy creative fiction\""
      },
      {
        "title": "Private-only mode (zero data retention, no Big Tech proxying)",
        "body": "python3 {baseDir}/scripts/venice-router.py --private-only --prompt \"Analyze this confidential contract\""
      },
      {
        "title": "Conversation-aware routing (multi-turn context)",
        "body": "# Save conversation history as JSON, then route follow-ups with context\npython3 {baseDir}/scripts/venice-router.py --conversation history.json --prompt \"Can you add tests too?\"\n\nThe router analyzes conversation history to keep context: trivial follow-ups (\"thanks\") go cheap, while follow-ups in complex code discussions stay at the right tier."
      },
      {
        "title": "Function calling (tool use)",
        "body": "# Define tools in a JSON file (OpenAI tools format)\npython3 {baseDir}/scripts/venice-router.py --tools tools.json --prompt \"What's the weather in NYC?\"\npython3 {baseDir}/scripts/venice-router.py --tools tools.json --tool-choice auto --prompt \"Search for latest AI news\"\n\nTool definitions use the standard OpenAI format. The router auto-bumps to mid tier minimum for function calling since it requires capable models."
      },
      {
        "title": "Cost budget tracking",
        "body": "# Show current spending\npython3 {baseDir}/scripts/venice-router.py --budget-status\n\n# Track per-session costs\npython3 {baseDir}/scripts/venice-router.py --session-id my-project --prompt \"help me code\"\n\nSet VENICE_DAILY_BUDGET and/or VENICE_SESSION_BUDGET to enforce spending limits. The router auto-downgrades tiers as you approach budget limits."
      },
      {
        "title": "Classify only (no API call)",
        "body": "python3 {baseDir}/scripts/venice-router.py --classify \"Explain the Riemann hypothesis\""
      },
      {
        "title": "List available models and tiers",
        "body": "python3 {baseDir}/scripts/venice-router.py --list-models"
      },
      {
        "title": "Override model directly",
        "body": "python3 {baseDir}/scripts/venice-router.py --model deepseek-v3.2 --prompt \"Hello\""
      },
      {
        "title": "Tiers",
        "body": "TierModelsCost (input/output per 1M tokens)Best ForcheapVenice Small (qwen3-4b), GLM 4.7 Flash, GPT OSS 120B, Llama 3.2 3B$0.05–$0.15 / $0.15–$0.60Simple Q&A, greetings, math, lookupsbudgetQwen 3 235B, Venice Uncensored, GLM 4.7 Flash Heretic$0.14–$0.20 / $0.75–$0.90Moderate questions, summaries, translationsbudget-mediumGrok Code Fast, DeepSeek V3.2, MiniMax M2.1$0.25–$0.40 / $1.00–$1.87Moderate-to-complex tasks, code snippets, structured outputmidDeepSeek V3.2, MiniMax M2.1/M2.5, Qwen3 Thinking 235B, Venice Medium, Llama 3.3 70B$0.25–$0.70 / $1.00–$3.50Code generation, analysis, longer writing, reasoninghighGLM 5, Kimi K2 Thinking, Kimi K2.5, Grok 4.1 Fast, Hermes 3 405B, Gemini 3 Flash$0.50–$1.10 / $1.25–$3.75Complex reasoning, multi-step tasks, code reviewpremiumGPT-5.2, GPT-5.2 Codex, Gemini 3 Pro, Gemini 3.1 Pro (1M ctx), Claude Opus/Sonnet 4.5/4.6$2.19–$6.00 / $15.00–$30.00Expert-level analysis, architecture, research papers"
      },
      {
        "title": "Routing Strategy",
        "body": "The router classifies each prompt using keyword + heuristic analysis:\n\nLength — longer prompts suggest more complex tasks\nKeywords — domain-specific terms (e.g., \"architecture\", \"optimize\", \"prove\") signal complexity\nCode markers — presence of code blocks, function names, or technical syntax\nInstruction depth — multi-step instructions, comparisons, or \"explain in detail\" bump the tier\nConversational simplicity — greetings, yes/no, small talk stay on the cheapest tier\nConversation history — when --conversation is provided, analyzes full chat context: code in history boosts tier, trivial follow-ups (\"thanks\") downgrade, tool calls in history signal complexity\nFunction calling — --tools auto-bumps to at least mid tier (capable models required)\nThinking/reasoning mode — --thinking prefers chain-of-thought reasoning models (Qwen3 Thinking, Kimi K2) and bumps to at least mid tier\nBudget constraints — progressive tier downgrade as spending approaches daily/session limits (95% → cheap, 80% → budget, 60% → mid, 40% → high)\n\nThe classifier errs on the side of cheaper models — it only escalates when there's strong signal for complexity."
      },
      {
        "title": "Environment Variables",
        "body": "VariableDescriptionDefaultVENICE_API_KEYVenice.ai API key (required)—VENICE_DEFAULT_TIERMinimum floor tier — auto-classification never goes below this. Valid: cheap, budget, budget-medium, mid, high, premiumbudgetVENICE_MAX_TIERMaximum tier to ever use (cost cap)premiumVENICE_TEMPERATUREDefault temperature0.7VENICE_MAX_TOKENSDefault max tokens4096VENICE_STREAMEnable streaming by defaultfalseVENICE_UNCENSOREDAlways prefer uncensored modelsfalseVENICE_PRIVATE_ONLYOnly use private models (zero data retention)falseVENICE_WEB_SEARCHEnable web search by default ($10/1K calls)falseVENICE_THINKINGAlways prefer thinking/reasoning modelsfalseVENICE_DAILY_BUDGETMax daily spend in USD (0 = unlimited)0VENICE_SESSION_BUDGETMax per-session spend in USD (0 = unlimited)0"
      },
      {
        "title": "Why Venice.ai?",
        "body": "🔒 Private inference — Models marked \"Private\" have zero data retention. Your data never trains anyone's model.\n🔓 Uncensored — No guardrails blocking legitimate use cases. No refusals, no filters.\n🔌 OpenAI-compatible — Same API format, just change the base URL. Drop-in replacement.\n📦 30+ models — From tiny efficient models ($0.05/M) to Claude Opus 4.6 and GPT-5.2.\n🌐 Built-in web search — LLMs can search the web and cite sources in a single API call."
      },
      {
        "title": "Tips",
        "body": "Use --classify to preview which tier a prompt would hit before spending tokens\nSet VENICE_MAX_TIER=mid to cap costs and never hit premium models\nUse --uncensored for creative, security research, or other content mainstream AI won't touch\nUse --private-only when processing sensitive/confidential data — zero retention guaranteed\nUse --web-search when you need up-to-date information with cited sources\nUse --conversation with a JSON message history for smarter multi-turn routing\nUse --tools to enable function calling — the router auto-bumps to capable models\nSet VENICE_DAILY_BUDGET=1.00 to cap daily spend at $1 — the router auto-downgrades tiers as you approach the limit\nUse --budget-status to see a detailed breakdown of your spending by tier\nUse --thinking for math proofs, logic puzzles, and multi-step reasoning — routes to Qwen3 Thinking or Kimi K2 models\nThe router prefers private (self-hosted) Venice models over anonymized ones when available at the same tier\nWhen --uncensored is active, the router auto-bumps to the nearest tier with uncensored models\nCombine with OpenClaw WebChat for a seamless chat experience routed through Venice.ai"
      }
    ],
    "body": "Venice.ai Supreme Router\n\nSmart, cost-optimized model routing for Venice.ai — the AI platform for people who don't want Big Tech watching over their shoulder.\n\nUnlike OpenAI, Anthropic, and Google — where every prompt is logged, analyzed, and potentially used to train future models — Venice offers true privacy with zero data retention on private models. Your conversations stay yours. Venice is also uncensored: no content filters, no refusals, no \"I can't help with that.\"\n\nSetup\nGet a Venice.ai API key from venice.ai/settings/api\nSet the environment variable:\nexport VENICE_API_KEY=\"your-key-here\"\n\n\nOr configure in ~/.openclaw/openclaw.json:\n\n{\n  \"skills\": {\n    \"entries\": {\n      \"venice-router\": {\n        \"enabled\": true,\n        \"apiKey\": \"YOUR_VENICE_API_KEY\"\n      }\n    }\n  }\n}\n\nUsage\nRoute a prompt (auto-selects model)\npython3 {baseDir}/scripts/venice-router.py --prompt \"What is 2+2?\"\n\nForce a specific tier\npython3 {baseDir}/scripts/venice-router.py --tier cheap --prompt \"Tell me a joke\"\npython3 {baseDir}/scripts/venice-router.py --tier budget-medium --prompt \"Write a Python function\"\npython3 {baseDir}/scripts/venice-router.py --tier mid --prompt \"Explain quantum computing\"\npython3 {baseDir}/scripts/venice-router.py --tier premium --prompt \"Write a distributed systems architecture\"\n\nStream output\npython3 {baseDir}/scripts/venice-router.py --stream --prompt \"Write a poem about lobsters\"\n\nWeb search (LLM searches the web and cites sources)\npython3 {baseDir}/scripts/venice-router.py --web-search --prompt \"Latest news on AI regulation\"\n\nUncensored mode (prefer models with no content filters)\npython3 {baseDir}/scripts/venice-router.py --uncensored --prompt \"Write edgy creative fiction\"\n\nPrivate-only mode (zero data retention, no Big Tech proxying)\npython3 {baseDir}/scripts/venice-router.py --private-only --prompt \"Analyze this confidential contract\"\n\nConversation-aware routing (multi-turn context)\n# Save conversation history as JSON, then route follow-ups with context\npython3 {baseDir}/scripts/venice-router.py --conversation history.json --prompt \"Can you add tests too?\"\n\n\nThe router analyzes conversation history to keep context: trivial follow-ups (\"thanks\") go cheap, while follow-ups in complex code discussions stay at the right tier.\n\nFunction calling (tool use)\n# Define tools in a JSON file (OpenAI tools format)\npython3 {baseDir}/scripts/venice-router.py --tools tools.json --prompt \"What's the weather in NYC?\"\npython3 {baseDir}/scripts/venice-router.py --tools tools.json --tool-choice auto --prompt \"Search for latest AI news\"\n\n\nTool definitions use the standard OpenAI format. The router auto-bumps to mid tier minimum for function calling since it requires capable models.\n\nCost budget tracking\n# Show current spending\npython3 {baseDir}/scripts/venice-router.py --budget-status\n\n# Track per-session costs\npython3 {baseDir}/scripts/venice-router.py --session-id my-project --prompt \"help me code\"\n\n\nSet VENICE_DAILY_BUDGET and/or VENICE_SESSION_BUDGET to enforce spending limits. The router auto-downgrades tiers as you approach budget limits.\n\nClassify only (no API call)\npython3 {baseDir}/scripts/venice-router.py --classify \"Explain the Riemann hypothesis\"\n\nList available models and tiers\npython3 {baseDir}/scripts/venice-router.py --list-models\n\nOverride model directly\npython3 {baseDir}/scripts/venice-router.py --model deepseek-v3.2 --prompt \"Hello\"\n\nTiers\nTier\tModels\tCost (input/output per 1M tokens)\tBest For\ncheap\tVenice Small (qwen3-4b), GLM 4.7 Flash, GPT OSS 120B, Llama 3.2 3B\t$0.05–$0.15 / $0.15–$0.60\tSimple Q&A, greetings, math, lookups\nbudget\tQwen 3 235B, Venice Uncensored, GLM 4.7 Flash Heretic\t$0.14–$0.20 / $0.75–$0.90\tModerate questions, summaries, translations\nbudget-medium\tGrok Code Fast, DeepSeek V3.2, MiniMax M2.1\t$0.25–$0.40 / $1.00–$1.87\tModerate-to-complex tasks, code snippets, structured output\nmid\tDeepSeek V3.2, MiniMax M2.1/M2.5, Qwen3 Thinking 235B, Venice Medium, Llama 3.3 70B\t$0.25–$0.70 / $1.00–$3.50\tCode generation, analysis, longer writing, reasoning\nhigh\tGLM 5, Kimi K2 Thinking, Kimi K2.5, Grok 4.1 Fast, Hermes 3 405B, Gemini 3 Flash\t$0.50–$1.10 / $1.25–$3.75\tComplex reasoning, multi-step tasks, code review\npremium\tGPT-5.2, GPT-5.2 Codex, Gemini 3 Pro, Gemini 3.1 Pro (1M ctx), Claude Opus/Sonnet 4.5/4.6\t$2.19–$6.00 / $15.00–$30.00\tExpert-level analysis, architecture, research papers\nRouting Strategy\n\nThe router classifies each prompt using keyword + heuristic analysis:\n\nLength — longer prompts suggest more complex tasks\nKeywords — domain-specific terms (e.g., \"architecture\", \"optimize\", \"prove\") signal complexity\nCode markers — presence of code blocks, function names, or technical syntax\nInstruction depth — multi-step instructions, comparisons, or \"explain in detail\" bump the tier\nConversational simplicity — greetings, yes/no, small talk stay on the cheapest tier\nConversation history — when --conversation is provided, analyzes full chat context: code in history boosts tier, trivial follow-ups (\"thanks\") downgrade, tool calls in history signal complexity\nFunction calling — --tools auto-bumps to at least mid tier (capable models required)\nThinking/reasoning mode — --thinking prefers chain-of-thought reasoning models (Qwen3 Thinking, Kimi K2) and bumps to at least mid tier\nBudget constraints — progressive tier downgrade as spending approaches daily/session limits (95% → cheap, 80% → budget, 60% → mid, 40% → high)\n\nThe classifier errs on the side of cheaper models — it only escalates when there's strong signal for complexity.\n\nEnvironment Variables\nVariable\tDescription\tDefault\nVENICE_API_KEY\tVenice.ai API key (required)\t—\nVENICE_DEFAULT_TIER\tMinimum floor tier — auto-classification never goes below this. Valid: cheap, budget, budget-medium, mid, high, premium\tbudget\nVENICE_MAX_TIER\tMaximum tier to ever use (cost cap)\tpremium\nVENICE_TEMPERATURE\tDefault temperature\t0.7\nVENICE_MAX_TOKENS\tDefault max tokens\t4096\nVENICE_STREAM\tEnable streaming by default\tfalse\nVENICE_UNCENSORED\tAlways prefer uncensored models\tfalse\nVENICE_PRIVATE_ONLY\tOnly use private models (zero data retention)\tfalse\nVENICE_WEB_SEARCH\tEnable web search by default ($10/1K calls)\tfalse\nVENICE_THINKING\tAlways prefer thinking/reasoning models\tfalse\nVENICE_DAILY_BUDGET\tMax daily spend in USD (0 = unlimited)\t0\nVENICE_SESSION_BUDGET\tMax per-session spend in USD (0 = unlimited)\t0\nWhy Venice.ai?\n🔒 Private inference — Models marked \"Private\" have zero data retention. Your data never trains anyone's model.\n🔓 Uncensored — No guardrails blocking legitimate use cases. No refusals, no filters.\n🔌 OpenAI-compatible — Same API format, just change the base URL. Drop-in replacement.\n📦 30+ models — From tiny efficient models ($0.05/M) to Claude Opus 4.6 and GPT-5.2.\n🌐 Built-in web search — LLMs can search the web and cite sources in a single API call.\nTips\nUse --classify to preview which tier a prompt would hit before spending tokens\nSet VENICE_MAX_TIER=mid to cap costs and never hit premium models\nUse --uncensored for creative, security research, or other content mainstream AI won't touch\nUse --private-only when processing sensitive/confidential data — zero retention guaranteed\nUse --web-search when you need up-to-date information with cited sources\nUse --conversation with a JSON message history for smarter multi-turn routing\nUse --tools to enable function calling — the router auto-bumps to capable models\nSet VENICE_DAILY_BUDGET=1.00 to cap daily spend at $1 — the router auto-downgrades tiers as you approach the limit\nUse --budget-status to see a detailed breakdown of your spending by tier\nUse --thinking for math proofs, logic puzzles, and multi-step reasoning — routes to Qwen3 Thinking or Kimi K2 models\nThe router prefers private (self-hosted) Venice models over anonymized ones when available at the same tier\nWhen --uncensored is active, the router auto-bumps to the nearest tier with uncensored models\nCombine with OpenClaw WebChat for a seamless chat experience routed through Venice.ai"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/PlusOne/venice-router",
    "publisherUrl": "https://clawhub.ai/PlusOne/venice-router",
    "owner": "PlusOne",
    "version": "1.5.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/venice-router",
    "downloadUrl": "https://openagent3.xyz/downloads/venice-router",
    "agentUrl": "https://openagent3.xyz/skills/venice-router/agent",
    "manifestUrl": "https://openagent3.xyz/skills/venice-router/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/venice-router/agent.md"
  }
}