agent-framework-bedrock

What it is and what it does

agent-framework-bedrock is a provider package that bridges Agent Framework applications to Amazon Bedrock's large language models. It wraps Bedrock's API behind the Agent Framework's chat abstractions, letting you send conversation turns and attach tools for function calling using the framework's standard ChatOptions interface.

The package depends on agent-framework-core, boto3, and botocore. It is in beta but actively maintained, with support for Python 3.10 through 3.14. You configure it via environment variables and instantiate BedrockChatClient to start making requests. The integration is designed for developers already using Agent Framework who want to route their agent workloads through Bedrock models.

Use it for:

• Route Agent Framework applications to Bedrock models instead of other LLM providers
• Build agents with tool calling on Bedrock by attaching tools through ChatOptions
• Migrate existing Agent Framework code to use Bedrock as the underlying model backend
• Test agent behavior across multiple Bedrock model families from a single codebase

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Connects Microsoft Agent Framework applications to Amazon Bedrock models, enabling chat interactions and tool/function calling through a unified interface.

Yes, if you are already using Microsoft Agent Framework and want to use Amazon Bedrock models. The package has low install friction, active maintenance, no known vulnerabilities, and a permissive MIT license. It is in beta (version 1.0.0b260730), so expect potential API changes before a stable release.

Install

agent-framework-bedrock on PyPI

pip install agent-framework-bedrock

uv add agent-framework-bedrock

poetry add agent-framework-bedrock

Installing agent-framework-bedrock

Before you install

Low install friction with three lightweight runtime dependencies (agent-framework-core, boto3, botocore). Active maintenance with a recent release 15 days old and ongoing repository activity.

License in practice

MIT license permits broad use, modification, and distribution with minimal restrictions.

Quickstart

pip install agent-framework-bedrock --pre

from agent_framework_bedrock import BedrockChatClient

client = BedrockChatClient()
response = client.send_message("Hello")

Requires AWS credentials via BEDROCK_* environment variables and Python 3.10 or later.

Verify before relying

• Whether BedrockChatClient supports all Bedrock model families or a specific subset
• Specific tool/function calling capabilities and limitations compared to direct Bedrock API
• Authentication flow details beyond BEDROCK_* environment variables

Package facts

Evidence: agent_framework_bedrock-1.0.0b260730-py3-none-any.whl

Tags

More Application Frameworks packages

Further reading