Expose a Subset of ENA REST Services as MCP

The goal of this project is to provide a Model Context Protocol (MCP) server that is suitable for production and exposes expected REST endpoints from the European Nucleotide Archive (ENA) as organised, schema-driven tools for AI systems. Although ENA offers robust REST APIs for biological data access, direct integration with AI agents frequently results in inconsistent queries, low reproducibility, and lack of validation. In order to overcome this, the project implements an MCP-based interface layer that ensures predictable execution, provides a strict input/output schemas, and enables secure, secure communication with ENA services. A dynamic tool registry for flexibility, Pydantic for schema validation, httpx for asynchronous API communication, and a response normalisation layer for consistent outputs are all features of the FastMCP-built system. To ensure scalability and production readiness, it will also include a comprehensive error handling, testing, and containerised deployment using Docker and Kubernetes. A fully functional MCP server with several ENA capabilities (such as study, sample, and run queries), thorough test coverage, developer-friendly documentation, and a deployable system that is simple to interface with AI clients are some of the key deliverables. This project will enable more dependable AI-driven scientific workflows by improving the reproducibility, traceability, and accessibility of biological data.