# Send Smart Router to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- 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.
## Suggested prompts
### New install

```text
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.
```
### Upgrade existing

```text
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.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "smart-models",
    "name": "Smart Router",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/samstone908/smart-models",
    "canonicalUrl": "https://clawhub.ai/samstone908/smart-models",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/smart-models",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=smart-models",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "README.md",
      "SKILL.md",
      "models.json",
      "scripts/call-model.sh",
      "scripts/sync-models.sh"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "slug": "smart-models",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-03T19:06:57.155Z",
      "expiresAt": "2026-05-10T19:06:57.155Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=smart-models",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=smart-models",
        "contentDisposition": "attachment; filename=\"smart-models-1.0.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "smart-models"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/smart-models"
    },
    "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."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/smart-models",
    "downloadUrl": "https://openagent3.xyz/downloads/smart-models",
    "agentUrl": "https://openagent3.xyz/skills/smart-models/agent",
    "manifestUrl": "https://openagent3.xyz/skills/smart-models/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/smart-models/agent.md"
  }
}
```
## Documentation

### Smart Router — Intelligent Model Router

Route tasks to the best model automatically, via any OpenAI-compatible API.

Author: whatevername2023@proton.me

### Setup

Models and provider are configured in models.json. Set two environment variables:

SMART_ROUTER_BASE_URL — OpenAI-compatible API base URL (e.g. https://api.openai.com/v1)
SMART_ROUTER_API_KEY — API key for the provider

Edit models.json to customize categories, models, and defaults for your provider.

### @ Alias Shortcuts

Prefix a message with @alias to skip auto-classification and call a specific model directly.

Format: @alias your question or prompt here

### Alias Table

AliasModel IDCategoryVision@gpt4ochatgpt-4o-latestvision@qwen-vlqwen3-vl-235b-a22b-instructvision@qwen-vl-maxqwen-vl-max-2025-08-13vision@llama-vlllama-3.2-90b-vision-instructvision@qwen-vl-32bqwen3-vl-32b-instructvisionImage Gen@imagengoogle/imagen-4-ultraimage_gen@fluxblack-forest-labs/flux-1.1-pro-ultraimage_gen@flux-kontextblack-forest-labs/flux-kontext-maximage_gen@dalledall-e-3image_gen@flux2flux-2-proimage_genVideo Gen@sorasora-2-pro-allvideo_gen@veoveo3.1-pro-4kvideo_gen@viduviduq3-provideo_gen@klingkling-videovideo_gen@runwayrunwayml-gen4_turbo-10video_genAudio@sunosuno_musicaudio@ttsgemini-2.5-pro-preview-ttsaudio@tts-hdtts-1-hdaudio@kling-audiokling-audioaudio@vidu-ttsvidu-ttsaudioReasoning@o3o3reasoning@o3-proo3-proreasoning@o4-minio4-minireasoning@deepseekdeepseek-r1reasoning@gemini-thinkgemini-2.5-pro-thinkingreasoning@claude-thinkclaude-sonnet-4-5-20250929-thinkingreasoningCode@claudeclaude-opus-4-6code@codexgpt-5.1-codex-maxcode@claude-sonnetclaude-sonnet-4-6code@qwen-coderqwen3-coder-480b-a35b-instructcode@qwen-coder-plusqwen3-coder-pluscode@gpt4tgpt-4-turbocodeGeneral@gpt52 / @gpt5gpt-5.2-chat-latestgeneral@geminigemini-2.5-progeneral@deepseekv3deepseek-v3.2general@qwenqwen3-maxgeneral@claude-chatclaude-opus-4-6general

Aliases are case-insensitive. If no alias matches, attempt fuzzy match on model name/ID. If still no match, prompt the user.

### Auto-Classification Rules

When no @alias is specified, classify the task automatically:

CategoryTriggervisionUser sends image/URL, asks to analyze, describe, OCR, understand image contentimage_genRequests to draw, generate image, design poster, create illustrationvideo_genRequests to generate video, animation, text-to-video, image-to-videoaudioRequests for music generation, TTS, sound effectsreasoningComplex math, logic puzzles, proofs, deep analysis, long-chain reasoningcodeCode generation, debugging, refactoring, review (when external model needed)generalEveryday chat, translation, summarization, writing, Q&A

### 1. Read Model Config

cat "$(dirname "$0")/../models.json"

### 2. Select Model

Determine category based on classification rules above
Use the first model with "default": true in each category
If user specifies a model via @alias, use that model directly
For cost-sensitive tasks, pick a smaller model in the same category

### 3. Call Model

Chat (vision / reasoning / code / general)

scripts/call-model.sh --model "MODEL_ID" --prompt "user request" --type chat

With image (vision):

scripts/call-model.sh --model "MODEL_ID" --prompt "request" --type chat --image "IMAGE_URL"

Image Generation

scripts/call-model.sh --model "MODEL_ID" --prompt "image description" --type image

Async Tasks (video / audio)

scripts/call-model.sh --model "MODEL_ID" --prompt "task description" --type async

TTS

scripts/call-model.sh --model "MODEL_ID" --prompt "text to speak" --type tts --voice alloy

### 4. Return Results

Chat: return the model's text reply directly
Image: return the generated image URL in markdown format
Video/Audio: return task status and result URL

### Model Recommendations

Vision: qwen3-vl-235b-a22b-instruct (strongest visual understanding)
Image gen: google/imagen-4-ultra (highest quality)
Video: sora-2-pro-all (best results)
Music: suno_music / TTS: tts-1-hd or gemini-2.5-pro-preview-tts
Reasoning: o3 (strongest reasoning)
Code: gpt-5.1-codex-max
General: claude-opus-4-6

### Fallback

If a model call fails, automatically fall back to the next model in the same category.

### Customization

Edit models.json to:

Add/remove models in any category
Change default models
Add new categories
Update aliases in SKILL.md to match

The scripts/sync-models.sh script lists all available models from your provider to help discover new ones.
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: samstone908
- Version: 1.0.0
## Source health
- Status: healthy
- Item download looks usable.
- Yavira can redirect you to the upstream package for this item.
- Health scope: item
- Reason: direct_download_ok
- Checked at: 2026-05-03T19:06:57.155Z
- Expires at: 2026-05-10T19:06:57.155Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/smart-models)
- [Send to Agent page](https://openagent3.xyz/skills/smart-models/agent)
- [JSON manifest](https://openagent3.xyz/skills/smart-models/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/smart-models/agent.md)
- [Download page](https://openagent3.xyz/downloads/smart-models)