---

name: docling description: Extract and parse content from web pages, PDFs, documents (docx, pptx), and images using the docling CLI with GPU acceleration. Use INSTEAD of web_fetch for extracting content from specific URLs when you need clean, structured text. Use Brave (web_search) for searching/discovering pages. Use docling when you HAVE a URL and need its content parsed. version: 1.0.2 metadata: requires: bins: ["docling"]

Docling - Document & Web Content Extraction

CLI tool for parsing documents and web pages into clean, structured text. Uses GPU acceleration for OCR and ML models.

Prerequisites

• docling CLI must be installed (e.g., via pipx install docling)
• For GPU support: NVIDIA GPU with CUDA drivers

When to Use

• Extract content from a URL → Use docling (not web_fetch)
• Search for information → Use web_search (Brave)
• Parse PDFs, DOCX, PPTX → Use docling
• OCR on images → Use docling

Quick Commands

Web Page → Markdown (default)

docling "<URL>" --from html --to md

Output: creates a .md file in current directory (or use --output)

Web Page → Plain Text

docling "<URL>" --from html --to text --output /tmp/docling_out

PDF with OCR

docling "/path/to/file.pdf" --ocr --device cuda --output /tmp/docling_out

Key Options

Option	Values	Description
--from	html, pdf, docx, pptx, image, md, csv, xlsx	Input format
--to	md, text, json, yaml, html	Output format
--device	auto, cuda, cpu	Accelerator (default: auto)
--output	path	Output directory (recommended: use controlled temp dir)
--ocr	flag	Enable OCR for images/scanned PDFs
--tables	flag	Extract tables (default: on)

Security Notes

⚠️ Avoid these flags unless you trust the source:

• --enable-remote-services - can send data to remote endpoints
• --allow-external-plugins - loads third-party code
• Custom --headers with untrusted values - can redirect requests

Workflow

1. For web content extraction: Use docling "<URL>" --from html --to text --output /tmp/docling_out
2. Read the output file from the specified output directory
3. Clean up the output directory after reading

GPU Support

Docling supports GPU acceleration via CUDA (NVIDIA). Verify CUDA is available:

python -c "import torch; print(torch.cuda.is_available())"

Full CLI Reference

See references/cli-reference.md for complete option list.

Skill Library MCP

15,000+ ready-to-use skills for AI coding assistants, served on demand via MCP.

An MCP server that provides on-demand skill loading for AI coding assistants. Instead of stuffing your system prompt with every skill you might need, this server indexes 15,000+ skills and serves only the ones relevant to your current task — keeping context windows lean and responses focused.

Documentation

Full documentation is at modbender.in/skill-library-mcp — installation, the tools it exposes, configuration, and examples.

Why?

• 15,000+ skills covering frontend, backend, DevOps, security, testing, databases, AI/ML, automation, and more
• On-demand loading — skills are fetched only when needed, not crammed into every conversation
• IDF-weighted search — finds the right skill even from natural language queries like "help me debug a memory leak"
• Browse by category — 13 categories to discover skills you didn't know existed
• Works with any MCP-compatible tool — Claude Code, Cursor, Windsurf, VS Code, Claude Desktop, and others
• Claude Code plugin — one-command install with claude plugin install
• Zero config — run with npx, no setup needed

Quick Start

Claude Code Plugin (Recommended)

Add the marketplace source, then install the plugin:

claude plugin marketplace add https://github.com/modbender/skill-library-mcp.git --scope user
claude plugin install skill-library --scope user

The MCP server starts automatically when Claude Code launches. No manual configuration needed.

Claude Code (MCP Server)

claude mcp add skill-library --scope user -- npx -y skill-library-mcp

MCP Server (Other Tools)

Add to your claude_desktop_config.json (location varies by OS):

{
  "mcpServers": {
    "skill-library": {
      "command": "npx",
      "args": ["-y", "skill-library-mcp"]
    }
  }
}

Add to .cursor/mcp.json (project) or ~/.cursor/mcp.json (global):

{
  "mcpServers": {
    "skill-library": {
      "command": "npx",
      "args": ["-y", "skill-library-mcp"]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "skill-library": {
      "command": "npx",
      "args": ["-y", "skill-library-mcp"]
    }
  }
}

Add to .vscode/mcp.json:

{
  "servers": {
    "skill-library": {
      "command": "npx",
      "args": ["-y", "skill-library-mcp"]
    }
  }
}

git clone https://github.com/modbender/skill-library-mcp
cd skill-library-mcp
pnpm install
pnpm build

Then point your MCP config to the built binary:

{
  "mcpServers": {
    "skill-library": {
      "command": "node",
      "args": ["/path/to/skill-library-mcp/dist/index.js"]
    }
  }
}

Tools

search_skill

Search for skills by keyword. Returns a ranked list of matching skill names and descriptions.

search_skill({ query: "react patterns" })

load_skill

Load the full content of a skill by name. Optionally includes resource files.

load_skill({ name: "brainstorming", include_resources: true })

list_categories

Browse all skill categories with counts and examples. Use to discover skills before searching.

list_categories()

Skill Categories

The library includes 15,000+ skills across 13 categories:

Category	Examples
Frontend	React patterns, Angular, Vue, Svelte, Next.js, Tailwind, accessibility
Backend	Node.js, FastAPI, Django, NestJS, Express, GraphQL, REST API design
AI & LLM	LLM app dev, RAG implementation, agent patterns, prompt engineering, embeddings
DevOps & Infra	Terraform, Kubernetes, Docker, AWS, GCP, Azure, CI/CD
Data & Databases	PostgreSQL, MongoDB, Redis, SQL optimization, ETL pipelines, analytics
Security	Penetration testing, OWASP, threat modeling, vulnerability scanning, encryption
Testing	TDD workflows, Playwright, Vitest, Jest, E2E testing patterns
Mobile	React Native, Flutter, iOS, Android, Expo
Automation	Workflow automation, n8n, Zapier, web scraping, bots
Python	Django, Flask, FastAPI, pandas, Python tooling
TypeScript & JS	TypeScript, JavaScript, Deno, Bun
Architecture	Microservices, system design, design patterns, monorepos
Other	Hundreds of specialized and niche skills

Skill Format

Skills are directories containing a SKILL.md file with YAML frontmatter:

---
name: my-skill
description: What this skill does
---

# My Skill

Skill content here...

Skills can optionally include a resources/ directory with additional .md files that are appended when include_resources: true is set.

Contributing

Contributions are welcome! To add a new skill:

1. Create a directory under data/ with your skill name
2. Add a SKILL.md file with YAML frontmatter (name, description)
3. Run pnpm dedup to check for duplicates
4. Submit a PR

Development

pnpm install          # Install dependencies
pnpm test             # Run tests
pnpm build            # Build to dist/
pnpm dev              # Run server locally
pnpm dedup            # Check for duplicate skills
pnpm validate-skills  # Validate data/ directory structure
pnpm fix-skills       # Fix broken skills (dry run by default)
pnpm clean-skills     # Remove invalid skill dirs (dry run by default)
make ci               # Run test + validate + build

Third-Party Content

This project includes skills from openclaw/skills, licensed under the MIT License. See THIRD_PARTY_NOTICES.md for details.

License

MIT