loading…
Search for a command to run...
loading…
Exposes the structure and documentation of the openreview-py library, enabling LLMs to discover classes, functions, and signatures for accurate code generation.
Exposes the structure and documentation of the openreview-py library, enabling LLMs to discover classes, functions, and signatures for accurate code generation. It provides tools for searching the API and retrieving detailed metadata without executing code or making external API calls.
MCP server that helps LLMs write correct openreview-py code. Two knowledge layers: live introspection of the installed library (method signatures, docstrings, class structures) and static knowledge (best practices, code examples, workflow guides).
docker build -t openreview-mcp /path/to/openreview-mcp
Create a .mcp.json file in the root of the project you're working in (or in ~/.claude/ for global access):
{
"mcpServers": {
"openreview": {
"command": "docker",
"args": ["run", "--rm", "-i", "openreview-mcp"]
}
}
}
Knowledge files (llm.txt, examples.md) are bundled inside the image — no bind-mount required. To override with a live openreview-py checkout, add -e OPENREVIEW_KNOWLEDGE_PATH=/knowledge -v /path/to/openreview-py:/knowledge to the docker run args.
Restart Claude Code (/exit and relaunch). The 5 tools will be available immediately.
Add this to claude_desktop_config.json and restart Claude Desktop.
{
"mcpServers": {
"openreview-python-mcp-server": {
"command": "npx",
"args": []
}
}
}