vision_analyze
Analyze an image using a vision LLM to identify scenes, objects, lighting, clarity, and legible text. Provide a local file path or data URL to receive a detailed Chinese analysis report.
Instructions
Bare-API vision analysis of an image via vision-LLM (Qwen3-VL-235B-A22B-Instruct on <LLM_ENDPOINT_QWEN>). For image recognition / scene analysis / reading legible text identifiers. Pass either image_path (local file) or image_data_url (data:...;base64) — not both. Returns a Chinese analysis report. NOTE: this is a scene/text-visibility analyzer, NOT a barcode decoder — high-density barcode VALUES must be decoded by a real decoder (ZXing/Dynamsoft), the VLM cannot. Key read at runtime from opencode.json qwen provider.
Tool call template
This tool declares no required parameters. Add optional parameters from the input schema as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {},
"name": "vision_analyze"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| model | No | Vision model. Default Qwen3-VL-235B-A22B-Instruct. | |
| prompt | No | Custom analysis prompt. Default asks for a 5-part Chinese report (scene / objects / lighting-composition / clarity-occlusion / text, transliterate any visible text). | |
| image_path | No | Absolute path to a local image (.jpg/.png/.jpeg/.webp). Pass this OR image_data_url (one of the two required). | |
| timeout_sec | No | Request timeout. Default 60. | |
| image_data_url | No | Data URL: data:image/jpeg;base64,<...>. Pass this OR image_path (one of the two required). |