a2aproject/a2a-python
a2aproject/a2a-python has 2,226 GitHub stars and 533 forks. GitHot measured 0 stars in 24 hours and +5 in 7 days. Last checked 2026-11-04 23:06 UTC. Last push 2026-11-04.
Official Python SDK for the Agent2Agent (A2A) Protocol
What you can do
- A2A Protocol Compliant: Build agentic applications that adhere to the Agent2Agent (A2A) Protocol.
- Extensible: Easily add support for different communication protocols and database backends.
- Asynchronous: Built on modern async Python for high performance.
- HTTP servers (FastAPI, Starlette)
Measured star growth
- in 24h
- 0
- in 7 days
- +5
- in 15 days
- Update delayed
- in 1 month
- Update delayed
Net star changes over overlapping periods, not daily totals. Missing measurements are left blank. Stars measure attention, not software quality.
README.md
OriginalA2A Python SDK
A Python library for running agentic applications as A2A Servers, following the Agent2Agent (A2A) Protocol.
[!IMPORTANT] Upgrading the SDK from
0.3to1.0? See the v0.3 → v1.0 migration guide. For supported A2A spec versions, see Compatibility.
✨ Features
- A2A Protocol Compliant: Build agentic applications that adhere to the Agent2Agent (A2A) Protocol.
- Extensible: Easily add support for different communication protocols and database backends.
- Asynchronous: Built on modern async Python for high performance.
- Optional Integrations: Includes optional support for:
- HTTP servers (FastAPI, Starlette)
- gRPC
- OpenTelemetry for tracing
- SQL databases (PostgreSQL, MySQL, SQLite)
🧩 Compatibility
This SDK implements the A2A Protocol Specification 1.0, with compatibility mode for 0.3. See #742 for details on the compatibility scope.
| Spec Version | Transport | Client | Server |
|---|---|---|---|
1.0 |
JSON-RPC | ✅ | ✅ |
1.0 |
HTTP+JSON/REST | ✅ | ✅ |
1.0 |
gRPC | ✅ | ✅ |
0.3 (compat) |
JSON-RPC | ✅ | ✅ |
0.3 (compat) |
HTTP+JSON/REST | ✅ | ✅ |
0.3 (compat) |
gRPC | ✅ | ✅ |
🚀 Getting Started
Prerequisites
- Python 3.10+
uv(recommended) orpip
🔧 Installation
Install the core SDK and any desired extras using your preferred package manager.
| Feature | uv Command |
pip Command |
|---|---|---|
| Core SDK | uv add a2a-sdk |
pip install a2a-sdk |
| All Extras | uv add "a2a-sdk[all]" |
pip install "a2a-sdk[all]" |
| HTTP Server | uv add "a2a-sdk[http-server]" |
pip install "a2a-sdk[http-server]" |
| FastAPI Integration | uv add "a2a-sdk[fastapi]" |
pip install "a2a-sdk[fastapi]" |
| gRPC Support | uv add "a2a-sdk[grpc]" |
pip install "a2a-sdk[grpc]" |
| OpenTelemetry Tracing | uv add "a2a-sdk[telemetry]" |
pip install "a2a-sdk[telemetry]" |
| Encryption | uv add "a2a-sdk[encryption]" |
pip install "a2a-sdk[encryption]" |
| Database Drivers | ||
| PostgreSQL | uv add "a2a-sdk[postgresql]" |
pip install "a2a-sdk[postgresql]" |
| MySQL | uv add "a2a-sdk[mysql]" |
pip install "a2a-sdk[mysql]" |
| SQLite | uv add "a2a-sdk[sqlite]" |
pip install "a2a-sdk[sqlite]" |
| All SQL Drivers | uv add "a2a-sdk[sql]" |
pip install "a2a-sdk[sql]" |
Examples
Helloworld Example
Run Remote Agent
git clone https://github.com/a2aproject/a2a-samples.git cd a2a-samples/samples/python/agents/helloworld uv run .In another terminal, run the client
cd a2a-samples/samples/python/agents/helloworld uv run test_client.pyYou can validate your agent using the agent inspector. Follow the instructions at the a2a-inspector repo.
🌐 More Examples
You can find a variety of more detailed examples in the a2a-samples repository:
🤝 Contributing
Contributions are welcome! Please see the CONTRIBUTING.md file for guidelines on how to get involved.
📄 License
This project is licensed under the Apache 2.0 License. See the LICENSE file for more details.
Explore related projects
affaan-m/ECC
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
297,596 starsNousResearch/hermes-agent
The agent that grows with you
260,759 starsmultica-ai/andrej-karpathy-skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
229,722 starsfirecrawl/firecrawl
Supercharge your AI agents with data from the web and beyond. Building the library for superintelligence. 🔥
198,759 stars