{"id":"darkroaster-pubmearch","name":"pubmearch","af_score":51.0,"security_score":50.5,"reliability_score":22.5,"what_it_does":"Provides an MCP server (Python) that queries PubMed using NCBI advanced search, saves result files, and performs analyses such as keyword hotspot frequency, keyword trends over time, publication count over time windows, and generates comprehensive reports.","best_when":"You want an AI-agent-accessible MCP tool to explore PubMed research dynamics (hotspots/trends/publication counts) using NCBI advanced search syntax.","avoid_when":"You need strong guarantees around reliability (SLA, documented retries/idempotency) or you cannot provide/secure NCBI credentials and want strict credential-less operation.","last_evaluated":"2026-03-30T13:42:02.993224+00:00","has_mcp":true,"has_api":false,"auth_methods":["Environment variables: NCBI_USER_EMAIL","Environment variables: NCBI_USER_API_KEY"],"has_free_tier":false,"known_gotchas":["Tool inputs rely on correctly formatted PubMed advanced search syntax; malformed queries may fail or return unexpected results.","Result files are saved to a local directory (pubmearch/results); agents may need to manage filesystem state and avoid reusing/overwriting prior runs.","No explicit documentation provided for pagination semantics beyond max_results; large queries may be truncated.","No explicit retry/idempotency guidance documented for long-running or partial failures."],"error_quality":0.0}