AWS Bedrock Data Automation — Claude Code MCP server setup
AWS Bedrock Data Automation — an MCP server to inspect documents, images, video recordings, and audio files via Amazon Bedrock.
Apache-2.0 ★ 9,939 updated 2026-10-23
Open on GitHubWhat AWS Bedrock Data Automation does
Part of the awslabs/mcp collection, this server links your AI assistant to automated media processing in AWS. It enables queries against text files, scanned paperwork, imagery, audio, and video files directly from supported coding environments.
Operations run through Amazon Bedrock endpoints, which means valid AWS credentials and permissions are necessary beforehand. The maintainers mention Agent Toolkit for AWS as the broader evolution of these utilities, while existing servers in the repository remain functional.
What it helps with
- extract structured insights from incoming scanned documents
- inspect graphics and photographs using Bedrock processing pipelines
- process voice recordings and video clips directly within agent sessions
Example prompts for the agent
- Inspect the document
contract.pdfwith AWS Bedrock Data Automation and list the main clauses - Send
interview.mp3to Bedrock for automated data extraction - Evaluate
recording.mp4to locate key events and their timestamps
How to install AWS Bedrock Data Automation
Install
The MCP registry has no launch command for it: see the repository README
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about AWS Bedrock Data Automation
How to install AWS Bedrock Data Automation in Claude Code?
Configure it inside your .mcp.json settings file using the server run command from the awslabs/mcp repository.
What does AWS Bedrock Data Automation do?
It gives coding agents tools to process and understand documents, still images, audio, and video files via Amazon Bedrock.
What is required before using this server?
You need an AWS environment configured with access to Amazon Bedrock and working authentication credentials.