# visual-analysis-ocr

Reads text from PNG images, maps document structure, and turns the result into markdown. Useful for scans and visual documents.

- subagent · `type=agent` · Claude Code · 742 installs
- Author: davila7
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/davila7/claude-code-templates/tree/HEAD/cli-tool/components/agents/ocr-extraction-team/visual-analysis-ocr.md
- Updated: 2026-10-24
- HTML: https://affpapa.org/en/nevibecode/agents/visual-analysis-ocr

## What it does

This subagent reads PNG images and extracts their text in reading order. It can identify headings, lists, captions, footnotes, symbols, mathematical notation, columns, indentation, emphasis, quotes, code blocks, tables, and diagram labels.

The output uses markdown headings, list markers, emphasis, code formatting, and preserved paragraph breaks. Ambiguous passages, difficult layouts, and non-text elements are flagged or described. Its configured tools are `Read` and `Write`, so the workflow centers on reading the image and writing the analysis.

Install it in Claude Code with `npx claude-code-templates@latest --agent ocr-extraction-team/visual-analysis-ocr`, then use `.claude/agents/visual-analysis-ocr.md`.

## Install

### Claude Code (native platform)

1. **claude-code-templates**
```bash
npx claude-code-templates@latest --agent ocr-extraction-team/visual-analysis-ocr
```

2. **Manually: put the files in place**
- File: `.claude/agents/visual-analysis-ocr.md` — For all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention

### Codex

**Manually: put the files in place**
- File: `.codex/agents/visual-analysis-ocr.toml` — Codex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/

### Cursor

**Manually: put the files in place**
- File: `.cursor/agents/visual-analysis-ocr.md` — Cursor also reads .claude/agents/

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/davila7/claude-code-templates/tree/HEAD/cli-tool/components/agents/ocr-extraction-team/visual-analysis-ocr.md

## When to use it
- extract text from a PNG
- rebuild heading and list structure
- convert image content into markdown
- read text across multiple columns
- flag unclear passages and diagram elements

## Example prompts
- Extract the text from this PNG and keep its reading order in markdown.
- Turn the headings, lists, and emphasis in this image into a markdown structure.
- Inspect the image, flag unclear sections, and describe its tables and diagram labels.

## Similar
- [vision-analysis](https://affpapa.org/en/nevibecode/skills/vision-analysis.md) — skill · 2700 installs
- [deepseek-ocr](https://affpapa.org/en/nevibecode/skills/deepseek-ocr.md) — skill · 1600 installs
- [PDF OCR Extraction](https://affpapa.org/en/nevibecode/skills/pdf-ocr-extraction.md) — skill · 5700 installs
- [silicon-paddle-ocr](https://affpapa.org/en/nevibecode/skills/silicon-paddle-ocr.md) — skill · 1200 installs
- [ocr-super-surya](https://affpapa.org/en/nevibecode/skills/ocr-super-surya.md) — skill · 558 installs
- [gemini-image](https://affpapa.org/en/nevibecode/skills/gemini-image.md) — skill · 1200 installs

From the same set: [davila7/claude-code-templates](https://affpapa.org/en/nevibecode/by/davila7/claude-code-templates.md) — 495 cards

---
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