Skip to content
Documentation
Cookbook
Multimodal: Vision, Charts, and Transcription
Crop tool for image analysis

Crop tool for image analysis

Adds a crop tool to let Claude zoom into regions of interest, enabling a practical agentic loop for detailed image analysis.

  • Upstream notebook: multimodal/crop_tool.ipynb

What to focus on

  • The “discover → crop → re-analyze” loop
  • Tool design (coordinates, outputs) and safety
  • When cropping beats longer prompts

Run locally 

make test-notebooks NOTEBOOK=multimodal/crop_tool.ipynb
Transcribe documentsUsing a sub-agent (Haiku) for extraction