Short answer: Drop the "Anthropic → Analyze Image" action anywhere in your workflow, map the inputs from upstream nodes, and publish.
Every field can be mapped from an upstream trigger, AI step, table row, or hard-coded literal.
| Field | Type | Required | Description |
|---|---|---|---|
Model model | options | Required | Model. Options: Claude Sonnet 4.6, Claude Opus 4.6, Claude Haiku 4.5 |
Image (base64) image_data | string | Required | Base64-encoded image data (PNG, JPEG, GIF, or WebP) |
Image Format media_type | options | Optional | Image Format. Options: PNG, JPEG, GIF, WebP |
Prompt prompt | string | Required | Prompt. e.g. "What do you see in this image?" |
Max Tokens max_tokens | string | Optional | Max Tokens |
{"model": "{{trigger.model}}","image_data": "{{trigger.image_data}}","media_type": "{{trigger.media_type}}","prompt": "What do you see in this image?","max_tokens": "{{trigger.max_tokens}}"}
{"id": "msg_123","type": "message","model": "claude-sonnet-4-6-20250514","usage": {"input_tokens": 1200,"output_tokens": 150},"content": [{"text": "The image shows a golden retriever sitting on a green lawn...","type": "text"}]}
Use these fields in downstream nodes for routing, logging, or error handling.
Any of these apps can fire this action as part of a workflow.
Triggered by anything in the catalog. Free tier available. No credit card.