Biolab
Auditable API and MCP infrastructure for agent-facing retrieval
A production-style service that exposes external data through MCP while preserving provenance, rate limits, source responses, and tamper-evident audit records.
AI agents can query scientific databases without leaving enough evidence to reconstruct the exact source response, query, and retrieval context behind a later claim.
Exposes search_pubmed, search_europepmc, search_clinicaltrials, search_biorxiv, and get_retrieval over streamable HTTP MCP.
Stores verbatim source responses and structured snapshots in one source-agnostic audit schema backed by SQLite locally or Turso in production.
Chains each retrieval hash to the previous record and serializes writes through one background writer so tampering or deletion can be detected.
Uses per-caller API keys and isolated rate-limit budgets while keeping a low-throughput anonymous tier available.
Publicly hosted on Hugging Face Spaces, backed by Turso, listed in the official MCP Registry and Smithery, distributed through PyPI, GHCR, GitHub Releases, and Go binaries/packages.