---
id: agent-framework-claude
version: "1.0.0b260813"
license: unclear
license_treatment: permissive
maintenance: active
---

# agent-framework-claude — Claude Agent SDK integration for Microsoft Agent Framework.

License: permissive · Maintenance: active · Downloads: 331.6K/mo

## What it is and what it does

Agent-framework-claude is a Claude-specific integration layer for Microsoft Agent Framework, enabling you to build production-grade AI agents and multi-agent workflows in Python using Claude as your LLM provider. It bridges Claude with the broader Agent Framework ecosystem, which provides orchestration patterns (sequential, concurrent, handoff, group collaboration), middleware support, observability via OpenTelemetry, and workflow management features like checkpointing and human-in-the-loop control.

The package depends on agent-framework-core and claude-agent-sdk, and is designed for teams moving agents from prototype to production. It supports Python 3.10 through 3.14 and is currently in Beta status. The framework emphasizes provider flexibility so your architecture can evolve without major rewrites as requirements change.

Use it for:

- Build a multi-agent customer support system where Claude agents handle different ticket categories with handoff and escalation workflows.
- Create a research assistant that orchestrates multiple Claude agents to gather, analyze, and synthesize information from different sources.
- Develop a production chatbot with durability, observability, and human-in-the-loop approval gates using Agent Framework's workflow patterns.
- Implement a group collaboration workflow where multiple Claude agents debate or refine solutions through structured orchestration.
- Deploy agents to Foundry-hosted infrastructure with minimal code changes while maintaining local development and testing capability.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Integrates Claude with Microsoft Agent Framework to enable building production-grade AI agents and multi-agent workflows in Python using Claude as the LLM provider.

Yes, with conditions. Install if you are building production agents with Claude and need orchestration beyond simple chat loops, or if you want to leverage Agent Framework's multi-agent patterns, observability, and provider flexibility. The package is actively maintained, has no known vulnerabilities, and uses a permissive MIT license. Verify that your Claude credentials and the runtime dependencies are compatible with your deployment model before committing to production use.

## Install

pip install agent-framework-claude
uv add agent-framework-claude
poetry add agent-framework-claude

## Installing agent-framework-claude

Before you install: Low install friction with a pure-Python wheel distribution. The package is actively maintained with a recent release and 12804 repository stars, indicating active development and community engagement.

License in practice: Licensed under MIT (permissive), allowing free use, modification, and distribution with minimal restrictions—suitable for both open-source and commercial projects.

Quickstart:

pip install agent-framework-claude

import asyncio
from agent_framework import Agent

async def main():
    agent = Agent(
        name="MyAgent",
        instructions="You are a helpful assistant.",
    )
    result = await agent.run("Your prompt here")
    print(result)

asyncio.run(main())

Requires Python 3.10 or later. Depends on agent-framework-core and claude-agent-sdk runtime dependencies.

Verify before relying:

- Whether this package requires a Claude API key or other external credentials to function
- Specific version compatibility between agent-framework-claude and its runtime dependencies agent-framework-core and claude-agent-sdk
- Whether the package includes documentation or examples beyond the Microsoft Agent Framework main repository

## Package facts

- License: not declared (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 331.6K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

claude agent framework python, multi-agent workflows claude, production ai agents microsoft, claude sdk agent orchestration, ai agent framework integration, claude llm provider agents, agent orchestration python, multi-agent-orchestration, claude-integration, production-agents

[View on SkillFed](https://skillfed.io/packages/agent-framework-claude) · [View on PyPI](https://pypi.org/project/agent-framework-claude/)