# Agent2Agent (A2A) Protocol

Agent2Agent (A2A) is an open protocol enabling communication and interoperability between opaque agentic applications.

Repository: https://github.com/a2aproject/A2A
Canonical: https://ross.abutalabs.com/products/agent2agent-a2a-protocol
Homepage: https://a2a-protocol.org/
Language: Shell
License: Apache-2.0
License Family: permissive
Topics: a2a, a2a-mcp, a2a-protocol, a2a-server, agents, generative-ai, linux-foundation
Last push: 2026-08-25T16:28:14+00:00
Link (homepage): https://a2a-protocol.org/

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 86, longevity 37
- inputs: {"age_days": 526, "days_push": 8, "days_rel": 97, "gap_med": 12.0, "n_releases_24m": 11}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 25503, forks 2586 (observed 2026-08-28T04:11:38.194420+00:00)

## What it is
Agent2Agent (A2A) is an open protocol, hosted under the Linux Foundation, that standardizes communication and interoperability between opaque agentic applications. It includes a protocol specification, a Python SDK (a2a-sdk), and sample implementations for building A2A-compatible agents and servers.

## Use cases
- make my ai agent talk to other agents
- standard protocol for agent-to-agent communication
- build an a2a-compatible agent server in python
- interoperate agents from different frameworks
- expose my agent as a service other agents can call
- replace custom agent integration glue with an open standard

## When to choose
- You need agents built on different frameworks or vendors to communicate without sharing internals
- You want a Linux Foundation-backed open standard with an official Python SDK and samples
- You are exposing an agent as a networked service discoverable by other agents

## When to avoid
- You only need a single agent calling tools within one process - MCP or a plain SDK is simpler
- You need agent-to-tool integration rather than agent-to-agent communication
- Your stack has no SDK support yet and implementing the protocol from spec is too costly

## Facets
- artifact type: library
- maturity: active
- function: agent-framework, rpc, api-framework, sdk
- domain: large-language-models, apis, artificial-intelligence
- platform: python, cross-platform
- tags: a2a-protocol, agent-interop, agent-communication, linux-foundation, protocol-specification, generative-ai, ai-agents

## Member repositories
- a2aproject/A2A (main) score 82
- a2aproject/a2a-python (sdk) score 86
- a2aproject/a2a-samples (examples) score 72

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:38.194420+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:56:02.857082+00:00, confidence not recorded.
  - readme: https://github.com/a2aproject/A2A (fetched 2026-08-28T04:11:38.194420+00:00, sha f4e4c3e6116a)
  - homepage: https://a2a-protocol.org/ (fetched 2026-08-29T07:52:48.171971+00:00, sha 36c6c3c2e4f9)
- Data as of 2026-08-30T08:39:29.467469+00:00.