# PaperPlotHub

An open, AI-reviewed marketplace where researchers share the plotting scripts behind their published figures, so others can find a template, its rendered output and its source paper side by side.

PaperPlotHub is a gallery of academic paper figures, each submission showing the original published figure, a reproduction rendered locally from an attached script, and the script itself, tagged by chart type, style and paper, with an arXiv link. Visitors can browse by chart type, view a paper's original and reproduced figure side by side, copy the script or download the rendered image with no login required.

Contributors sign in with GitHub and submit in one of three modes: a full submission with the original figure, reproduction and script; a reproduction-only submission from their own experiment referencing a paper's setup; or a paper-only submission flagging a figure worth reproducing for someone else to pick up. Every submission passes a lightweight AI review, a static security scan, arXiv metadata retrieval, and a vision-LLM check that the rendered figure actually came from the submitted script, before high-confidence submissions publish automatically.

Agents can submit through a stdio MCP server or a token-authenticated REST API; the site states it never executes an uploaded script itself.

Website: https://paperplothub.tech/