# pdf-to-markdown

pdf-to-markdown loads an entire PDF into the agent's context as structured Markdown with tables and images instead of grep-sized fragments.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Windsurf (Devin Desktop), Claude.ai · 14 installs · ★ 5 repo stars
- Author: aliceisjustplaying
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/aliceisjustplaying/claude-resources-monorepo/tree/HEAD/skills/pdf-to-markdown/SKILL.md
- Updated: 2025-12-22
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/aliceisjustplaying-pdf-to-markdown

## What it does

The skill runs IBM's Docling model to turn a whole PDF into Markdown: font size drives heading levels, the TableFormer model handles table extraction, and lists, multi-column layout and code blocks come through intact. Images get pulled out separately into an `images/` folder next to the resulting `.md` file, linked from the text with relative paths, and listed in a summary table at the bottom.

It runs inside its own environment at `~/.claude/skills/pdf-to-markdown/.venv/`, built with `uv` and the `pymupdf`, `docling` and `docling-core` packages on first use. Output gets cached by file content hash under `~/.cache/pdf-to-markdown/`, so repeat requests for the same file come back instantly, while the first pass takes longer because Docling has to pull down its models.

Scanned PDFs need Tesseract OCR installed, or text extraction quality drops. When the markdown output references a figure or chart, the agent is meant to open that image file with Read on its own rather than just repeating the caption.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add aliceisjustplaying/claude-resources-monorepo --skill pdf-to-markdown -a claude-code
```

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

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add aliceisjustplaying/claude-resources-monorepo --skill pdf-to-markdown -a codex
```

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

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add aliceisjustplaying/claude-resources-monorepo --skill pdf-to-markdown -a cursor
```

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

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install aliceisjustplaying/claude-resources-monorepo skills/pdf-to-markdown
```

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

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add aliceisjustplaying/claude-resources-monorepo --skill pdf-to-markdown -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/aliceisjustplaying/claude-resources-monorepo --path skills/pdf-to-markdown
```

### Windsurf (Devin Desktop)

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

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add aliceisjustplaying/claude-resources-monorepo --skill pdf-to-markdown -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/aliceisjustplaying/claude-resources-monorepo/tree/HEAD/skills/pdf-to-markdown/SKILL.md

## When to use it
- pull a full PDF into an agent's working context
- extract tables from a PDF without breaking their structure
- get images out of a PDF with links back to the surrounding text
- process a scanned document through OCR
- skip re-converting a PDF that was already processed

## Example prompts
- Load the whole document.pdf into context and summarize the main sections
- Convert report.pdf to Markdown with its tables and images
- Read paper.pdf in full and describe what the figures show

## Similar
- [markdown-converter](https://affpapa.org/en/nevibecode/skills/steipete-markdown-converter.md) — skill · 346 installs
- [PDF to Markdown](https://affpapa.org/en/nevibecode/mcp/pdf-to-markdown.md) — MCP server
- [oma-pdf](https://affpapa.org/en/nevibecode/skills/oma-pdf.md) — skill · ★ 1472 repo stars
- [pdf-toolkit-mcp](https://affpapa.org/en/nevibecode/mcp/pdf-toolkit-mcp.md) — MCP server · ★ 12 repo stars
- [pdf-mcp](https://affpapa.org/en/nevibecode/mcp/pdf-mcp.md) — MCP server · ★ 174 repo stars
- [pdf-extraction](https://affpapa.org/en/nevibecode/skills/pdf-extraction.md) — skill · 9600 installs

## Other versions with the same name
- [pdf-to-markdown · pspdfkit-labs](https://affpapa.org/en/nevibecode/skills/pdf-to-markdown.md) — 870 installs
- [pdf-to-markdown · claesbackman](https://affpapa.org/en/nevibecode/skills/claesbackman-pdf-to-markdown.md) — ★ 579 repo stars
- [pdf-to-markdown · lttr](https://affpapa.org/en/nevibecode/skills/lttr-pdf-to-markdown.md) — 5 installs

From the same set: [aliceisjustplaying/claude-resources-monorepo](https://affpapa.org/en/nevibecode/by/aliceisjustplaying/claude-resources-monorepo.md) — 2 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
