{"id":"oksure-openalex-research-mcp","name":"openalex-research-mcp","af_score":73.9,"security_score":50.8,"reliability_score":35.0,"what_it_does":"Provides an MCP (Model Context Protocol) server that exposes OpenAlex scholarly data to AI assistants via ~31 specialized tools for literature search, entity lookup, citation/reference analysis, trend analysis, and preset-based (journal/institution) constrained discovery. Includes CLI-based setup for MCP client integration (e.g., Claude Desktop) and supports options like caching, validation, and retry behavior.","best_when":"You want an MCP-compatible assistant to perform iterative literature search and analysis against OpenAlex data with structured tool access and preset-based constraints.","avoid_when":"You require a REST/GraphQL/SDK integration path or hard guarantees about response completeness/size for every tool call (summarized outputs are intentionally abbreviated).","last_evaluated":"2026-03-30T15:36:58.079823+00:00","has_mcp":true,"has_api":false,"auth_methods":["MCP server env var configuration: OPENALEX_EMAIL (optional)","Optional OPENALEX_API_KEY (for premium users)"],"has_free_tier":false,"known_gotchas":["Summarized list/search responses are intentionally truncated (e.g., authors limited to first 5 with a flag, and references/affiliations omitted) so follow-up calls like get_work may be required for completeness.","Some MCP clients (e.g., TypingMind) may produce 'tool_use_id' errors; troubleshooting suggests starting a new chat and requesting fewer results.","Pagination exists (page/per_page); agents requesting large result sets should manage per_page and pagination to avoid overly large responses."],"error_quality":0.0}