Retrieves information from the AWS Knowledge Base using the Bedrock Agent Runtime, enabling Retrieval-Augmented Generation (RAG).
The AWS Knowledge Base Retrieval tool is an MCP server implementation that facilitates retrieving information from an AWS Knowledge Base using the Bedrock Agent Runtime. It enables Retrieval-Augmented Generation (RAG) by retrieving context from the AWS Knowledge Base based on a user query and a specified Knowledge Base ID. The tool supports retrieving multiple results and returns both the raw context and structured RAG sources with metadata, allowing for flexible processing of the results. Configuration options include using IAM Access Keys or AWS SSO for authentication, and the ability to configure default Knowledge Base IDs.
Key Features
01RAG (Retrieval-Augmented Generation) for AWS Knowledge Base
02Supports retrieval of multiple results
03Returns raw context and structured RAG sources
04Supports IAM Access Keys and AWS SSO for authentication
05Optional configuration of default Knowledge Base IDs
Use Cases
01Integrating AWS Knowledge Base search into conversational AI applications
02Providing contextual information to large language models
03Building question answering systems powered by AWS Knowledge Base
