research-mcp
by Burton-David
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| NCBI_API_KEY | No | Optional API key for PubMed/NCBI to raise rate limit from 3 to 10 requests per second. | |
| OPENAI_API_KEY | No | API key for OpenAI, required if using OpenAI embedder (RESEARCH_MCP_EMBEDDER starts with 'openai') or analysis model (RESEARCH_MCP_ANALYSIS_MODEL starts with 'openai'). | |
| RESEARCH_MCP_EMBEDDER | No | Embedder to use for vector embeddings. Format: 'provider:model', e.g., 'openai:text-embedding-3-small' or 'sentence-transformers:BAAI/bge-base-en-v1.5'. Required for library ingest and recall. | |
| RESEARCH_MCP_RERANKER | No | Optional cross-encoder reranker for improved search quality. Format: 'cross-encoder:model', e.g., 'cross-encoder:BAAI/bge-reranker-base'. | |
| RESEARCH_MCP_INDEX_PATH | No | File path to store the FAISS index and metadata. Required for library tools (ingest, library_search, etc.). | |
| RESEARCH_MCP_DISABLE_PDF | No | Set to '1' to disable PDF full-text extraction during ingest even if pdfplumber is installed. | |
| SEMANTIC_SCHOLAR_API_KEY | No | Optional API key for Semantic Scholar to raise rate limits. | |
| RESEARCH_MCP_ANALYSIS_MODEL | No | LLM model for paper analysis (analyze_paper tool). Format: 'provider:model', e.g., 'openai:gpt-4o-mini' or 'anthropic:claude-haiku-4-5-20251001'. Required for analyze_paper. | |
| RESEARCH_MCP_DISABLE_PUBMED | No | Set to '1' to disable PubMed as a search source (default: enabled). | |
| RESEARCH_MCP_OPENALEX_EMAIL | No | Email address to identify with OpenAlex API (optional but good practice). | |
| RESEARCH_MCP_CITATION_SCORER | No | Optional LLM-based citation scorer to override the default heuristic scorer. Format: 'llm:provider:model', e.g., 'llm:anthropic:claude-haiku-4-5-20251001'. | |
| RESEARCH_MCP_CLAIM_EXTRACTOR | No | Optional LLM-based claim extractor to override the default spaCy extractor. Format: 'llm:provider:model', e.g., 'llm:anthropic:claude-haiku-4-5-20251001'. | |
| RESEARCH_MCP_S2_SHARED_RATELIMIT | No | Set to '1' to share Semantic Scholar rate limit across processes (POSIX only). Optional. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
No tools | |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
This server cannot be deployed
Maintenance
ActivityInactive
ResponsivenessSlow