S3 Tables MCP Server
An MCP server that lets agents discover, query, and reason over managed Apache Iceberg tables and S3 Metadata inventory tables in natural language under least-privilege access — no heavyweight external catalog required.
Summary
An MCP server that lets agents discover, query, and reason over managed Apache Iceberg tables and S3 Metadata inventory tables in natural language under least-privilege access — no heavyweight external catalog required.
The structured-data face of the Agentic Data Plane: it bridges lakehouse table formats and agentic reasoning, letting agents navigate multi-petabyte lakes by conversing with S3's automated metadata.
- It reasons over metadata and managed tables, not arbitrary objects — different scope from txn2/mcp-s3.
- "Least privilege" is load-bearing: agents see system properties/tags/events, not unrestricted data.
- S3 Tables MCP Server
extendsModel Context Protocol (MCP) - S3 Tables MCP Server
integrates_withAmazon S3 Tables, Apache Iceberg - Pairs with serverless engines (DuckDB) for laptop-scale lakehouse analytics
Definition
A specialized Model Context Protocol server that lets AI assistants discover, query, and derive insights from managed Apache Iceberg tables and S3 Metadata inventory tables using natural language instead of SQL — bridging structured lakehouse formats and agentic reasoning.
With billions of objects across enterprise AI lakes, querying metadata for lineage, access patterns, and compliance posture is a bottleneck. The server lets agents converse directly with S3's automated metadata (system properties, tags, events) under least-privilege access, without standing up a heavyweight external catalog.
Conversational metadata discovery, compliance/posture verification over multi-petabyte lakes, agent-initiated lifecycle management, lineage exploration.
Recent developments
- Least-privilege conversational access to S3 Metadata. Agents query system properties, custom tags, and event info directly. Per AWS Storage Blog — Derive intelligent storage insights using S3 Metadata and MCP.
- Pairs with serverless lakehouse query. Used alongside S3 Tables + engines like DuckDB for laptop-scale lakehouse analytics. Per Serverless Analytics from Your Laptop: S3 Tables, DuckDB, OpenAQ.
Connections 4
Outbound 4
scoped_to1extends1integrates_with2Resources 2
AWS's walkthrough of conversational insight over S3 Metadata and Iceberg tables via MCP — the least-privilege metadata-discovery pattern for agents.
Practical example pairing S3 Tables + DuckDB for laptop-scale lakehouse analytics, the engine context the MCP server sits alongside.