Scholar RAG MCP

Package (pypi 0.4.0)

uvx scholar-rag-mcp

Or add to your MCP client config:

{
  "mcpServers": {
    "scholar-rag-mcp": {
      "command": "uvx",
      "args": [
        "scholar-rag-mcp"
      ],
      "env": {
        "SCHOLAR_RAG_DATA_DIR": "<YOUR_VALUE>",
        "SCHOLAR_RAG_CHAT_BASE_URL": "<YOUR_VALUE>",
        "SCHOLAR_RAG_EMBED_BASE_URL": "<YOUR_VALUE>",
        "SCHOLAR_RAG_RERANK_BASE_URL": "<YOUR_VALUE>"
      }
    }
  }
}

• SCHOLAR_RAG_DATA_DIR — Data root directory (knowledge bases, caches, jobs)
• SCHOLAR_RAG_CHAT_BASE_URL required — OpenAI-compatible chat endpoint for annotation/keywords (e.g. local vLLM)
• SCHOLAR_RAG_EMBED_BASE_URL required — OpenAI-compatible embeddings endpoint
• SCHOLAR_RAG_RERANK_BASE_URL required — Rerank endpoint (/v1/rerank)