Amazon has launched an open-source Model Context Protocol (MCP) server that enables artificial intelligence (AI) assistants to search and preview datasets available through the Registry of Open Data on AWS, making it easier for researchers and developers to discover publicly available information.
‎‎The new MCP server connects compatible AI tools, including Kiro and Claude Code, to more than 1,100 datasets from over 400 organisations hosted on the Registry of Open Data on AWS.
‎‎The collections span satellite imagery, climate science, life sciences and geospatial data, with contributions from organisations such as NASA, NOAA, the NIH and the Allen Institute.
‎‎According to Amazon, the server allows users to interact with the datasets using natural language queries, helping them quickly identify relevant resources without manually searching through large data repositories.
‎‎In addition to dataset discovery, the tool enables users to explore metadata, access related resources and preview Amazon S3 bucket structures and file contents before downloading or conducting more detailed analysis.
‎‎Amazon said the MCP server is designed to streamline research workflows by allowing users to evaluate datasets before committing time and computing resources to deeper analysis.
‎‎The server offers three core capabilities: dataset discovery through natural language searches, exploration of metadata and similar resources, and evaluation through previews of file structures and content samples.
‎‎The company added that the server can be integrated with MCP-compatible AI assistants using uv configuration commands, allowing developers to incorporate the functionality into existing AI workflows.
‎‎Amazon has released the project as open-source software under the Apache 2.0 licence, with the source code now available on GitHub.
