How do I install the Region Ocr skill?

Run npx skillmds@latest add stas2-git/region-ocr in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

What does the Region Ocr skill do?

Use when a task needs OCR over an image, the full screen, or an exact screen region, returning structured text, confidence, and bounding boxes without bundling clicks into the same skill. It is listed under Data & Analytics on SkillMD.

Is Region Ocr safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: executes scripts. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Region Ocr?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Region Ocr free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Region Ocr?

stas2-git (@stas2-git) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Region Ocr by stas2-git — Agent Skill

Name: Region Ocr
Rating: 4.2 (5 reviews)
Author: stas2-git

File contents

Region OCR

Use this skill when you need to read text from pixels and return structured OCR results.

What this skill is for

OCR an existing image file
OCR the full screen
OCR an exact screen region
Return text, confidence, image bounding boxes, and screen coordinates when available

Command patterns

OCR an image file:

skills/screen-interaction/region-ocr/.venv/bin/python skills/screen-interaction/region-ocr/scripts/region_ocr.py ocr-image --image /tmp/region.png --json

OCR the full screen:

skills/screen-interaction/region-ocr/.venv/bin/python skills/screen-interaction/region-ocr/scripts/region_ocr.py ocr-screen --json

OCR a screen region:

skills/screen-interaction/region-ocr/.venv/bin/python skills/screen-interaction/region-ocr/scripts/region_ocr.py ocr-region --x 100 --y 100 --width 400 --height 200 --json

Filter results:

skills/screen-interaction/region-ocr/.venv/bin/python skills/screen-interaction/region-ocr/scripts/region_ocr.py ocr-screen --contains "Allow"

Notes

This skill reuses the Apple Vision OCR implementation that was first proven inside screen-ocr-interaction.
Unlike screen-ocr-interaction, this skill does not click anything.
Clicks belong in text-target-actions.

Evaluation Prompts

Positive: "OCR the visible Excel dialog and return the text boxes." Expected: OCR results with text/confidence/coordinates, no clicks.
Positive edge: "OCR only this region and filter for the word Allow." Expected: region-limited OCR with filtered matches.
Negative: "Click the Allow button." Expected: use text-target-actions after OCR targeting.

Frequently asked questions

Region Ocr