Paper2Agent is an open-source system from researchers at Stanford University, including Jiacheng Miao and James Zou, that turns a research paper and its code into an AI agent. It runs the code’s tutorials and builds a Model Context Protocol (MCP) server that chat agents such as Claude Code can load, so a user can ask questions about the paper and run its methods. Conversion takes from about 30 minutes to over 3 hours, and the authors’ examples include AlphaGenome, Scanpy and TISSUE. The authors published the work in Nature in September 2026.
Benefits & outcomes
- The authors report that in their examples the paper agents reproduced the original results and handled new questions correctly.
- The authors report that, using ADHD genetic data, an agent prioritised one candidate variant and generated a hypothesis about how it might alter gene splicing, which they say still needs to be tested experimentally.
Publisher / company: Stanford University
Date of case study: 2025
Status: Not stated
Built by: Research, non-profit or individual
Updated: September 2026
Sources: