image-analysis-requests

Guides the agent through submitting local image files and PDFs to Claude for visual analysis — confirming outbound transfers in advance, verifying the real file format, base64-encoding, resizing to manage token cost, and composing single-image, multi-image, few-shot, and document requests that return descriptions, transcribed text, table and chart readings, comparisons, and strict JSON. Use when the user wants a photo, screenshot, scan, diagram, or PDF analyzed, read, compared, or summarized, or when vision requests consume too many tokens. Not for creating or editing images, audio or video input, or hosted upload endpoints.

Version

v1.0.0

Size

15 KB

Updated

2026-10-13

Your client verifies the signature before reading. This explicit manual read is available on every plan. Free automatically loads five chosen skills; paid plans automatically load relevant skills across the library.