Amazon Bedrock Knowledge Base Slack Chat Bot

This project deploys a Slack ChatBot integration to a managed RAG service provided by Amazon Bedrock Knowledge base. All resources are deployed and managed using the AWS Cloud Development Kit (CDK). In this solution, an AWS API Gateway and AWS Lambda provide an interface to an Amazon Bedrock Knowledge base consisting of a vector database using AWS Opensearch serverless (AOSS) and an AOSS index via a custom resource.

The Slack integration is provided through the Slack Bolt Library for Python running in the Request Processor Lambda function. The Slack Bolt Library handles authentication and permissions to the Slack application. Slack Bolt provides a dedicated user guide to deploy and run the library in a Lambda function.

A detailed guide to deploy the project and the Slack App are given in the AWS blog: Create a generative AI assistant with Slack and Amazon Bedrock.

In this example, the Bedrock Knowledge base is populated with all of the public documentation of the AWS Well-Architected Framework. The SlackBot has been given a Slash Command function /ask-aws where users are able to ask any AWS design, architecture, security and other best practice related questions.

Managed Knowledge Bases (New)

This solution now supports Managed Knowledge Bases, which eliminate the need to provision and manage Amazon OpenSearch Serverless (AOSS).

With Managed Knowledge Bases:

• Bedrock handles embedding, storage, and retrieval automatically
• No AOSS collection required (simplified setup and reduced cost)
• Supports agentic retrieval with intelligent query decomposition and managed reranking
• Uses managedSearchConfiguration instead of vectorSearchConfiguration for the Retrieve API

The CDK stack has been updated to support both knowledge base types. Set the knowledge base type in cdk.json or via context:

cdk deploy --context knowledgeBaseType=MANAGED

When using a managed KB, the Lambda function automatically uses the correct retrieval configuration:

// Managed KB retrieval
retrievalConfiguration: {
  managedSearchConfiguration: {
    numberOfResults: 5
  }
}

SDK requirements: AWS SDK for JavaScript >= 3.750.0 for managed KB support.

Reranking options for managed search: MANAGED (default — automatic), NONE (disable reranking), CUSTOM (your own Bedrock reranking model e.g. Cohere Rerank v3.5).

Required IAM Permissions:

{
  "Effect": "Allow",
  "Action": [
    "bedrock:Retrieve",
    "bedrock:AgenticRetrieveStream"
  ],
  "Resource": "arn:aws:bedrock:<region>:<account-id>:knowledge-base/<kb-id>"
}

Resources: Build a Managed KB | Retrieve API | Agentic Retrieval

Giving Feedback and Contributions

• Contributions Guidelines Submit Issues, Feature Requests or Bugs

Code of Conduct

This project has adopted the Amazon Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opensource-codeofconduct@amazon.com with any additional questions or comments.

Security

See Security Issue Notifications for more information.

License

This library is licensed under the MIT-0 License. See the LICENSE file.