pdfdancer-mcp brings PDF editing SDK to AI coding assistants
`pdfdancer-mcp` is an MCP server by MenschMachine that connects the PDFDancer PDF editing SDK to MCP-compatible AI coding assistants, enabling them to write and refactor code for editing and transforming PDFs across TypeScript, Python, and Java projects.
Score breakdown
The server gives AI coding assistants live, searchable access to PDFDancer SDK documentation, enabling them to generate and refactor PDF-editing code for production workflows without requiring the developer to manually look up API references.
- 01MCP server connects the PDFDancer PDF editing SDK to AI coding assistants including Cursor, Claude Code, VS Code, Windsurf, and Zed.
- 02Exposes four tools: `help`, `version`, `search-docs`, and `get-docs`.
- 03`search-docs` returns up to 10 results with titles, content snippets, and relevance scores.
`pdfdancer-mcp`, published by MenschMachine, is a TypeScript-based remote MCP server that connects the PDFDancer PDF editing SDK to any MCP-compatible AI coding assistant. Rather than treating PDFs as read-only, PDFDancer exposes editing primitives at the level of paragraphs, lines, glyphs, vectors, forms, and fonts — including ML-driven handling for missing or embedded fonts — and `pdfdancer-mcp` makes those capabilities accessible directly inside an AI coding assistant's context.
Documentation queries are routed through a Cloudflare Worker that interfaces with a Docusaurus search service.
The server exposes four tools: `help` (a comprehensive SDK overview with multi-language code samples), `version` (current server version), `search-docs` (keyword search returning up to 10 matching routes with titles, snippets, and relevance scores), and `get-docs` (full markdown retrieval for a specific documentation route, including code examples and API references). Documentation queries are routed through a Cloudflare Worker that interfaces with a Docusaurus search service. Installation requires Node.js `>= v18.0.0` and can be done via a one-click button for Cursor or manually via `npx -y @pdfdancer/pdfdancer-mcp` configuration in any supported client. The server is positioned for use in PDF ingestion pipelines, document workflows, and production AI agents that need to modify inbound PDFs they did not originally create.
Key facts
- 01MCP server connects the PDFDancer PDF editing SDK to AI coding assistants including Cursor, Claude Code, VS Code, Windsurf, and Zed.
- 02Exposes four tools: `help`, `version`, `search-docs`, and `get-docs`.
- 03`search-docs` returns up to 10 results with titles, content snippets, and relevance scores.
- 04`get-docs` retrieves full markdown documentation including code examples and API references.
- 05Connects to a Cloudflare Worker backed by a Docusaurus search service for documentation queries.
- 06Supports TypeScript, Python, and Java projects.
- 07Requires Node.js >= v18.0.0; installable via `npx -y @pdfdancer/pdfdancer-mcp`.
Topics
Summary and scoring are generated automatically from the original article. We always link back to the publisher and never republish images or paywalled content. Last processed Jun 14, 2026 · 09:08 UTC. How this works →