Model Context Protocol

Published

Also known as: MCP, Model Context Protocol

An open standard that lets AI models connect to external tools and data sources — databases, APIs, file systems — through a universal interface without custom integration code.

Source: Anthropic (introduced November 2024)

Primary reference ↗

Model Context Protocol (MCP) is an open standard introduced by Anthropic in November 2024. It defines a uniform way for AI models to discover and invoke external tools — much like how USB provides a universal interface for hardware peripherals.

Problem MCP Solves

Before MCP, connecting an AI to a new data source (e.g., a protein database, an EHR system, a genomics API) required writing custom integration code for each combination of model and tool. With MCP, any MCP-compatible tool can be used by any MCP-compatible model without new integration work.

Relevance for Biological Research

An AI system built on MCP can connect to:

  • The Human Protein Atlas JSON API
  • Commercial antibody vendor databases
  • UniProt, Ensembl, PubMed
  • Local vector databases (Qdrant, Chroma)
  • Laboratory information systems (LIMS)

Each as a standardized “MCP server” — the AI model treats them all the same way regardless of the underlying data format.

MCP in Multi-Agent Systems

In a multi-agent biology system, MCP allows execution subagents to discover and call tools at runtime rather than having tool access hardcoded. This makes the architecture more modular and extensible as new data sources are added.