openreview
An MCP server for OpenReview to fetch user papers, specific conference papers (e.g. ICML / ICLR / NeurIPS), parse .pdf files and save them into .json. Can be added to Claude Code to sample ideas from the recent AI/ML/SWE conferences.
claude mcp add --transport stdio anyakors-openreview-mcp-server python -m openreview_mcp_server \ --env OPENREVIEW_BASE_URL="https://api2.openreview.net" \ --env OPENREVIEW_PASSWORD="your_password" \ --env OPENREVIEW_USERNAME="your_email@domain.com" \ --env OPENREVIEW_DEFAULT_EXPORT_DIR="./openreview_exports"
How to use
This MCP server provides access to OpenReview data for research and analysis. It exposes tools to search for OpenReview user profiles, retrieve papers authored by a given user, fetch conference papers from venues like ICML, ICLR, and NeurIPS, and perform keyword-based searches across multiple conferences. The server can also export results to JSON and PDF formats and can extract text from PDFs for downstream analysis. To use it, configure the openreview MCP server in your MCP client (Cursor or equivalent) and then issue natural language prompts or call the provided tools to perform searches, fetch papers, or export data. Typical workflows include searching for papers by keywords and venue/year, pulling all papers by a researcher, and exporting results for offline analysis or integration into other research pipelines.
How to install
Prerequisites:
- Python 3.12 or later
- Virtual environment tool (optional but recommended)
- Access to the internet to install dependencies
Step-by-step installation:
- Clone the repository
git clone https://github.com/yourusername/openreview-mcp-server.git
cd openreview-mcp-server
- Create and activate a virtual environment
python3 -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
- Install the package in editable mode
pip install -e .
- Prepare configuration (see mcp_config example above) and ensure required environment variables are set, e.g. in your MCP client configuration file
OPENREVIEW_USERNAME=your_email@example.com
OPENREVIEW_PASSWORD=your_password
OPENREVIEW_BASE_URL=https://api2.openreview.net
OPENREVIEW_DEFAULT_EXPORT_DIR=./openreview_exports
- Start or reload your MCP client so the openreview server becomes available via the configured path (the exact steps depend on your MCP client, e.g., Cursor).
Additional notes
Tips and common issues:
- Ensure you replace absolute paths in the mcp configuration with the actual path to your openreview-mcp-server directory.
- The server uses a Python virtual environment; make sure the path to the Python interpreter ends with /venv/bin/python (or equivalent on Windows).
- The OPENREVIEW_BASE_URL should point to the appropriate OpenReview API (e.g., https://api2.openreview.net).
- The default export directory will be created if it does not exist; ensure the process has write permissions to that directory.
- If you encounter authentication errors, double-check credentials and that your OpenReview account has the necessary access. Some endpoints may require additional permissions.
- When upgrading, reinstall in editable mode and restart your MCP client to pick up changes.
Related MCP Servers
mcp-vegalite
MCP server from isaacwasserman/mcp-vegalite-server
github-chat
A Model Context Protocol (MCP) for analyzing and querying GitHub repositories using the GitHub Chat API.
nautex
MCP server for guiding Coding Agents via end-to-end requirements to implementation plan pipeline
pagerduty
PagerDuty's official local MCP (Model Context Protocol) server which provides tools to interact with your PagerDuty account directly from your MCP-enabled client.
futu-stock
mcp server for futuniuniu stock
mcp -boilerplate
Boilerplate using one of the 'better' ways to build MCP Servers. Written using FastMCP