unstructured-pdf-generation

Generate synthetic PDF documents with JSON metadata for RAG testing.

27|9|Updated Feb 3, 2026
One-click install
npx skills add https://github.com/datasciencemonkey/claude-code-cli-bricks --skill unstructured-pdf-generation-datasciencemonkey
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: unstructured-pdf-generation
Source: https://github.com/datasciencemonkey/claude-code-cli-bricks/tree/main/.claude/skills/unstructured-pdf-generation
Command: npx skills add https://github.com/datasciencemonkey/claude-code-cli-bricks --skill unstructured-pdf-generation-datasciencemonkey

SYSTEM DOCUMENTATION & REQUIREMENTS

## What problem does it solve? This skill automates the creation of synthetic PDF documents for retrieval-augmented generation (RAG) workflows and unstructured data use cases, enabling rapid testing, demos, and evaluation datasets without exposing real data.

## Core Features & Use Cases

  • LLM-powered PDF generation: Create realistic PDFs with AI-generated content and structured metadata.
  • RAG-ready datasets: Produce accompanying JSON files with questions and evaluation guidelines for RAG testing.
  • Output management: Automatically save PDFs and JSONs to Unity Catalog Volumes under a specified catalog, schema, and folder.

### Quick Start Use the generate_pdf_documents MCP tool:

  • catalog: "my_catalog"
  • schema: "my_schema"
  • description: "Technical documentation for a cloud infrastructure platform including setup guides, troubleshooting procedures, and API references."
  • count: 10
  • volume: "raw_data"
  • folder: "pdf_documents"
  • overwrite_folder: true

Frequently Asked Questions about unstructured-pdf-generation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate synthetic PDFs for RAG testing?▼

You can generate synthetic PDFs for RAG testing by configuring a catalog, schema, description, and count to automatically produce AI-generated documents alongside JSON metadata for indexing.

What is the best way to create evaluation datasets for unstructured data workflows?▼

Creating evaluation datasets for unstructured data workflows is best handled by generating synthetic PDFs with accompanying JSON files containing questions and evaluation guidelines for RAG pipelines.

How do I save generated PDFs to Unity Catalog Volumes?▼

To save generated PDFs to Unity Catalog Volumes, specify the target catalog, schema, volume, and folder during generation, using the overwrite option to manage existing files.

Can I use LLM-generated documents for QA evaluation datasets without exposing real data?▼

Yes, you can use LLM-generated documents to create QA evaluation datasets, enabling rapid testing and demos for RAG pipelines without exposing real data.

Does synthetic PDF generation provide metadata suitable for RAG indexing?▼

Synthetic PDF generation provides JSON metadata suitable for indexing, ensuring the output documents are immediately ready for retrieval-augmented generation pipelines.

What do I need to configure to create demonstration samples for RAG pipelines?▼

To create demonstration samples for RAG pipelines, you need to configure the catalog, schema, description, and count, with optional volume and folder parameters for output management.