I've developed an MCP (Model Context Protocol) server that enables AI assistants like Claude and ChatGPT
to directly query ERDDAP oceanographic databases. The tool (
https://github.com/robertdcurrier/erddap2mcp)
allows users to search datasets, retrieve metadata, and analyze variables through natural language queries
rather than manual URL construction. Built on the official erddapy Python client, it supports both
tabledap and griddap protocols and includes data preview capabilities with summary statistics.
Installation is straightforward: clone the repository, run pip install -r requirements.txt, and add a
5-line JSON configuration to your AI assistant. The entire setup process takes under 5 minutes, and no
ERDDAP knowledge is required to start using it. The repository includes documentation with usage examples
demonstrating dataset discovery, variable analysis, and constrained data retrieval across multiple ERDDAP
servers including NOAA CoastWatch, IOOS, and GCOOS.
I'm including a screenshot of the MCP server in action with Claude Desktop. Feedback welcomed.
I'm sure there bugs that need to be squashed, but wanted to get this out there and let folks start beating on it.