AWS launched a Model Context Protocol (MCP) server for its Registry of Open Data on July 28, 2026. The server gives AI agents read-only access to over 170 public datasets via natural-language queries.
Key facts
- AWS launched MCP server for Registry of Open Data on July 28, 2026.
- Server provides read-only access to 170+ public datasets.
- MCP protocol crossed 13,000+ servers as of June 2026.
- Google launched its own MCP server for Google Cloud on July 8, 2026.
- MCP is backed by Anthropic, OpenAI, Block, Linux Foundation.
AWS launched a Model Context Protocol (MCP) server for its Registry of Open Data on July 28, 2026 According to the AWS blog post. The server provides read-only access to over 170 public datasets, including satellite imagery, genomic sequences, and climate models. Users can query datasets via natural language through MCP-compatible AI agents, bypassing the usual AWS Console navigation or manual download steps.
The Registry of Open Data on AWS has existed since 2018, hosting petabytes of public data at no cost to users (they only pay for compute). The MCP server effectively turns this catalog into a structured data source that AI agents can explore and retrieve from. For example, an AI agent could ask "Find satellite imagery of the Amazon basin from 2023" and receive a direct S3 path or file listing.
Why this matters
MCP is an open standard introduced by Anthropic in November 2024 and now backed by OpenAI, Block, and the Linux Foundation. As of June 2026, the protocol had crossed 13,000+ servers [per our prior reporting]. AWS's adoption is notable because it competes with Google Cloud and Microsoft Azure on AI infrastructure; Google launched its own MCP server for Google Cloud on July 8, 2026. AWS is the first of the big three to open its public-data registry via MCP.
The server is available now as an open-source MCP integration. AWS did not disclose adoption metrics or pricing for the MCP server usage (the data itself remains free). The move positions AWS's open-data catalog as a structured data source for AI agents, potentially reducing the friction of data discovery in AI workflows.
What to watch
Watch for Microsoft Azure to launch its own MCP server for its public-data catalog (e.g., Azure Open Datasets) in the next 60 days, and for AWS to disclose agent query volumes or new dataset integrations driven by the MCP server.
Source: news.google.com
