# databricks-unstructured-pdf-generation

databricks-unstructured-pdf-generation creates synthetic PDF files and paired evaluation questions locally, then copies them to Databricks Unity Catalog volumes for RAG testing.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Windsurf (Devin Desktop), Claude.ai · 328 installs · ★ 408 repo stars
- Author: databricks
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/databricks/databricks-agent-skills/tree/HEAD/skills/databricks-unstructured-pdf-generation/SKILL.md
- Updated: 2026-10-24
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/databricks-unstructured-pdf-generation

## What it does

This skill builds synthetic document collections for testing Databricks retrieval pipelines, such as Genie Agent setups or `mlflow.genai.evaluate()`. It writes domain-specific HTML pages into `./raw_data/html/` and converts them to PDF files using a bundled `pdf_generator.py` script that calls `plutoprint`.

Converted documents are copied into a Unity Catalog volume using `databricks fs cp`. The workflow also produces `pdf_eval_questions.json`, mapping each document to verification queries and expected facts.

Without `databricks CLI` version 1.0.0 or higher and the `plutoprint` library, conversions and volume sync will fail. Keep the evaluation JSON off the volume attached to a Genie Agent, because the agent indexes the entire volume and absorbs the test answers directly into its context.

## Requirements
- Works with: Databricks

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add databricks/databricks-agent-skills --skill databricks-unstructured-pdf-generation -a claude-code
```

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

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add databricks/databricks-agent-skills --skill databricks-unstructured-pdf-generation -a codex
```

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

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add databricks/databricks-agent-skills --skill databricks-unstructured-pdf-generation -a cursor
```

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

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install databricks/databricks-agent-skills skills/databricks-unstructured-pdf-generation
```

2. **Manually: put the files in place**
- File: `.github/skills/databricks-unstructured-pdf-generation/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add databricks/databricks-agent-skills --skill databricks-unstructured-pdf-generation -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/databricks/databricks-agent-skills --path skills/databricks-unstructured-pdf-generation
```

### Windsurf (Devin Desktop)

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

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add databricks/databricks-agent-skills --skill databricks-unstructured-pdf-generation -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/databricks/databricks-agent-skills/tree/HEAD/skills/databricks-unstructured-pdf-generation/SKILL.md

## When to use it
- produce synthetic multi-page PDFs from local HTML source files
- sync test PDF batches to Databricks Unity Catalog volumes
- build paired ground-truth question sets for RAG benchmarks
- prepare evaluation data for Genie Agent without leaking answers into its index

## Example prompts
- Generate HTML troubleshooting guides in `./raw_data/html` and convert them to PDF
- Copy local PDFs from `./raw_data/pdf` to volume `dbfs:/Volumes/my_catalog/my_schema/raw_data/pdf`
- Create `pdf_eval_questions.json` paired with the generated manual PDFs

## Similar
- [azure-databricks-platform-architect](https://affpapa.org/en/nevibecode/agents/azure-databricks-platform-architect.md) — subagent · ★ 6505 repo stars
- [databricks-cli](https://affpapa.org/en/nevibecode/plugins/databricks-cli.md) — plugin · ★ 1104 repo stars
- [databricks-expert](https://affpapa.org/en/nevibecode/skills/databricks-expert.md) — skill · 215 installs
- [data-engineering-patterns-fabric-databricks](https://affpapa.org/en/nevibecode/skills/data-engineering-patterns-fabric-databricks.md) — skill · 1600 installs
- [databricks-2025](https://affpapa.org/en/nevibecode/skills/databricks-2025.md) — skill · 168 installs
- [databricks](https://affpapa.org/en/nevibecode/rules/databricks.md) — rule · ★ 3589 repo stars

From the same set: [databricks/databricks-agent-skills](https://affpapa.org/en/nevibecode/by/databricks/databricks-agent-skills.md) — 30 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
