analyze_image
Send an image or multiple images to a vision model and get plain-text answers. Use when the host AI can't see images—ask questions, compare visuals, read text, or extract details with a custom prompt.
Instructions
Analyze one or more images with a vision model and return plain text. Use for general Q&A when the host model cannot see images. Supports single image or multi images (image 1, image 2, ...).
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": "analyze_image"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| image | No | Local file path, file:// URI, http(s) URL, data URL, or base64 image data | |
| images | No | One or more images. Prefer this for multi-image chats: ["path/a.png", "path/b.png"] or [{source, label: "1"}, {source, label: "2"}]. Labels default to "1", "2", ... | |
| prompt | No | What to analyze or answer (errors, UI review, chart meaning, differences, etc.) | |
| mimeType | No | Optional MIME type hint for a single bare-base64 `image` input, e.g. image/png |