#255 · Primary category: Education & Research

arxiv-mcp-server

ai arxiv claude-ai gpt llm mcp-server model-context-protocol papers python research

A local MCP server for agent literature work. Original-LaTeX section reads, BibTeX from arXiv metadata, and topic watches. Papers stay on disk. Search is optional.

Project last updated:08/26/26

GitHub Stars

3.1K

Forks

252

Contributors

18

License

Apache-2.0

Why we included this project

Anyone who does research through an AI assistant knows the friction of bouncing between the editor and the arXiv website to chase a citation or double-check a method. This MCP server removes that detour by letting the assistant query arXiv directly, with local stdio execution so papers and indexes stay on your machine. Beyond plain search it can download PDFs, read a bounded slice of full text, and pull the original LaTeX for a specific section, which matters when you want to inspect a proof or formula without wading through the whole paper. The citation graph is the most useful part: follow references forward and backward from one paper and trace how an idea evolved instead of tab-hopping across searches. Research alerts run locally as well, and if you use Claude Desktop, VS Code, or another MCP-capable client, this keeps the whole reading loop inside your tools.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category