# AWS S3 Tables

The AWS S3 Tables MCP server lets AI agents work with S3 Tables, run read-only SQL, inspect metadata, and turn CSV files in S3 into tables.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 9939 repo stars
- Author: awslabs
- License: Apache-2.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/awslabs/mcp/tree/HEAD/src/s3-tables-mcp-server
- Updated: 2026-10-23
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/aws-s3-tables

## What it does

AWS S3 Tables gives an agent tools for S3-based table storage. It can create and list table buckets, namespaces, and tables, rename tables, inspect schemas and storage metadata, and read bucket or table policies. It also supports read-only SQL queries and can create a table from a CSV file stored in S3.

The server starts in read-only mode. Creating resources or appending rows requires `--allow-write` together with suitable IAM permissions; SQL updates and deletes are not supported. Without configured AWS credentials and access to AWS services, the server cannot operate.

Install `uv`, then install Python with `uv python install 3.10`. Configure credentials with `aws configure` or environment variables. Add the server to the MCP client with `uvx` and `awslabs.s3-tables-mcp-server@latest`, setting `AWS_PROFILE` and `AWS_REGION`. Review agent permissions before enabling writes.

## Requirements
- Needs: Docker
- Works with: AWS

## Install

### Claude Code (native platform)

**Install**
- The MCP registry has no launch command for it: see the repository README

## When to use it
- inspect S3 table buckets and namespaces
- run read-only SQL against S3 Tables
- turn an S3 CSV file into a table
- read table schemas and storage metadata
- append new rows to a table

## Example prompts
- List the table buckets and namespaces available through AWS S3 Tables.
- Run a read-only SQL query against the S3 Tables and explain the result.
- Create an S3 Table from the CSV file stored in S3.

## Similar
- [querying-aws-s3](https://affpapa.org/en/nevibecode/skills/querying-aws-s3.md) — skill · 6400 installs
- [querying-aws-athena](https://affpapa.org/en/nevibecode/skills/querying-aws-athena.md) — skill · 454 installs
- [aws-cloud-services](https://affpapa.org/en/nevibecode/skills/aws-cloud-services.md) — skill · 452 installs
- [aws-s3](https://affpapa.org/en/nevibecode/skills/aws-s3.md) — skill · 382 installs
- [aws-cloudtrail](https://affpapa.org/en/nevibecode/skills/aws-cloudtrail.md) — skill · 376 installs
- [auditing-aws-s3-bucket-permissions](https://affpapa.org/en/nevibecode/skills/auditing-aws-s3-bucket-permissions.md) — skill · 307 installs

From the same set: [awslabs/mcp](https://affpapa.org/en/nevibecode/by/awslabs/mcp.md) — 36 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
