# i-am-bee/acp

Open protocol for communication between AI agents, applications, and humans.

Repository: https://github.com/i-am-bee/acp
Canonical: https://ross.abutalabs.com/products/acp
Homepage: https://agentcommunicationprotocol.dev
Language: Python
License: Apache-2.0
License Family: permissive
Topics: agents, protocol, python, typescript
Archived: true
Last push: 2025-08-25T18:42:36+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 38, release rhythm 44, longevity 36
- inputs: {"age_days": 517, "days_push": 373, "days_rel": 377, "gap_med": 0, "n_releases_24m": 46}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1018, forks 124 (observed 2026-08-28T04:03:14.929776+00:00)

## What it is
The Agent Communication Protocol (ACP) is an open protocol with Python and TypeScript SDKs for enabling AI agents built on different frameworks to communicate, coordinate, and discover each other via a RESTful API. It has been merged into the Agent2Agent (A2A) protocol under the Linux Foundation, so new adopters should use A2A.

## Use cases
- connect ai agents built with different frameworks
- make my agent discoverable over a rest api
- stream agent responses between services
- expose my agent as an mcp tool
- run long-running multi-agent tasks with shared state
- build an agent server in python or typescript

## When to choose
- you need agent-to-agent interoperability across frameworks like LangChain, CrewAI, or BeeAI
- you want a REST-based, runtime-free agent invocation protocol with streaming and session support
- you need an MCP adapter to expose ACP agents as MCP tools
- you want an open-governance protocol backed by the Linux Foundation

## When to avoid
- you are starting a new project - ACP has merged into A2A and is in maintenance
- you only need to connect a single LLM to tools and resources (use MCP instead)
- you need a full agent orchestration framework rather than a communication protocol

## Facets
- artifact type: library
- maturity: maintenance
- function: agent-framework, rpc, api-framework, sdk, mcp
- domain: large-language-models, apis, developer-tools
- platform: python, cross-platform
- tags: agent-communication-protocol, agent-interoperability, a2a, beeai, rest-api, multi-agent-systems, linux-foundation, merged-into-a2a, ai-agents, nodejs, docker

## Member repositories
- i-am-bee/acp (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:14.929776+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 2026-08-30T07:09:28.648479+00:00, confidence not recorded.
  - readme: https://github.com/i-am-bee/acp (fetched 2026-08-28T04:03:14.929776+00:00, sha 83e7927a59de)
  - homepage: https://agentcommunicationprotocol.dev (fetched 2026-08-29T13:09:56.678988+00:00, sha fe79fb907059)
  - site_page: https://agentcommunicationprotocol.dev/about/mission-and-team (fetched 2026-08-29T13:09:56.693380+00:00, sha 25b6047a91e6)
  - site_page: https://agentcommunicationprotocol.dev/about/contribute (fetched 2026-08-29T13:09:56.694912+00:00, sha 33e570838f65)
  - site_page: https://agentcommunicationprotocol.dev/about/whats-next (fetched 2026-08-29T13:09:56.696679+00:00, sha 086e78ad5d8c)
  - site_page: https://agentcommunicationprotocol.dev/about/mcp-and-a2a (fetched 2026-08-29T13:09:56.699152+00:00, sha e49b0240c917)
  - site_page: https://agentcommunicationprotocol.dev/introduction/quickstart (fetched 2026-08-29T13:09:56.688097+00:00, sha 96426debdecb)
  - site_page: https://agentcommunicationprotocol.dev/integrations/mcp-adapter (fetched 2026-08-29T13:09:56.690055+00:00, sha ac7a862c46a2)
  - site_page: https://agentcommunicationprotocol.dev/integrations/cursor (fetched 2026-08-29T13:09:56.691703+00:00, sha a1387e8d8ac3)
  - site_page: https://agentcommunicationprotocol.dev/introduction/whats-new (fetched 2026-08-29T13:09:56.701127+00:00, sha 49a0a07fcdcd)
- Data as of 2026-08-30T08:39:29.467469+00:00.