# analyze_image Extension

For full documentation, see the [project README](../README.md).

## Quick Start

```bash
# Install
mkdir -p ~/.pi/agent/extensions/analyze-image
ln -sf "$(pwd)/index.ts" ~/.pi/agent/extensions/analyze-image/index.ts

# Configure (optional)
cp config.example.json ~/.pi/agent/extensions/analyze-image/config.json
```

## Tool Parameters

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `images` | `string[]` | Yes | One or more local image file paths |
| `question` | `string` | Yes | Question about the image(s) |
| `model` | `string` | No | Override the configured default vision model |

## Config Fields

| Field | Type | Default | Description |
|-------|------|---------|-------------|
| `defaultModel` | string | `"gemma4:31b-cloud"` | Vision model for the subagent |
| `systemPrompt` | string | *(see source)* | System prompt for the subagent |
| `maxImagesPerCall` | number | `10` | Maximum images per call |

## Supported Formats

PNG, JPG, JPEG, GIF, WebP, BMP