# ocr

Extracts text from images and PDF files via optical recognition. It helps read tables from screenshots or digitise scanned documents directly in agent sessions.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 503 installs · ★ 2
- Author: mr-shaper
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/mr-shaper/opencode-skill-hybrid-ocr/tree/HEAD/SKILL.md
- Updated: 2026-02-04
- HTML: https://affpapa.org/en/nevibecode/skills/ocr

## What it does

The skill routes visual files to a local `scripts/ocr.py` Python runner. Standard execution relies on DeepSeek-OCR, while the `--fast` flag shifts processing to PaddleOCR for quicker results. Output can be emitted as raw text or structured via `--json`.

Custom instructions pass through the `--prompt` argument. This lets you ask the engine to output detected tables straight into Markdown. Python 3 must be available on your path to run the underlying script.

Supported image types cover PNG, JPG, JPEG, BMP, GIF, WEBP, and TIFF alongside standard PDF files. Over 100 languages are covered, with dedicated handling for Chinese and English text.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add mr-shaper/opencode-skill-hybrid-ocr --skill ocr -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/ocr/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/ocr/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add mr-shaper/opencode-skill-hybrid-ocr --skill ocr -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/ocr/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add mr-shaper/opencode-skill-hybrid-ocr --skill ocr -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/ocr/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install mr-shaper/opencode-skill-hybrid-ocr ocr
```

2. **Manually: put the files in place**
- File: `.github/skills/ocr/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add mr-shaper/opencode-skill-hybrid-ocr --skill ocr -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/mr-shaper/opencode-skill-hybrid-ocr
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/ocr/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add mr-shaper/opencode-skill-hybrid-ocr --skill ocr -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/mr-shaper/opencode-skill-hybrid-ocr/tree/HEAD/SKILL.md

## When to use it
- extract text from screenshot or photo assets
- convert printed tables on images into Markdown tables
- pull text layers from scanned PDF documents
- format optical character recognition results as structured JSON

## Example prompts
- Extract text from `document.png` and output the result as JSON.
- Parse `table.png` and render the table as markdown using `--prompt`.
- Run OCR on `invoice.pdf` in fast mode with the `--fast` flag.

## Similar
- [skill-creator](https://affpapa.org/en/nevibecode/skills/skill-creator.md) — skill · 414180 installs
- [writing-for-agents](https://affpapa.org/en/nevibecode/skills/writing-for-agents.md) — skill · 466719 installs
- [media-use](https://affpapa.org/en/nevibecode/skills/media-use.md) — skill · 1100000 installs
- [mcp-builder](https://affpapa.org/en/nevibecode/skills/mcp-builder.md) — skill · 130949 installs
- [subagent-driven-development](https://affpapa.org/en/nevibecode/skills/subagent-driven-development.md) — skill · 237746 installs
- [orchestration](https://affpapa.org/en/nevibecode/skills/orchestration.md) — skill · 350300 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md