agent-communication-protocol
Open protocol for AI agent interoperability enabling standardized communication between agents, applications, and humans across different frameworks
npx skills add majiayu000/claude-skill-registry --skill agent-communication-protocol-enuno-claude-command-and-c-2 --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
ACP defines an open protocol for interconnecting AI agents, applications, and humans with standardized communication across frameworks.
How it works
- It uses a REST-based design with HTTP endpoints (not JSON-RPC).
- It supports multi-modal content (text, images, audio, video, and any MIME type) and is async-first with optional synchronous and streaming (Server-Sent Events) modes.
- Sessions can be Stateful or Stateless, enabling some level of state across interactions.
- It includes an Agent Manifest for discovery and metadata, and provides REST API endpoints for discovery and run management (including runs, run status, and run events).
- The Quick Start demonstrates creating a simple Echo agent using an async generator that yields intermediate thoughts and messages.
- The Python SDK examples show client usage for discovery, asynchronous and streaming execution, and artifact handling.
When to use it
Use ACP when you need an open, framework-agnostic protocol to connect agents, apps, and humans across multiple environments with standardized message formats and discovery mechanisms.
What it can touch
- REST API endpoints (Discovery, Runs, Events)
- Message structures (Message, MessagePart) with content_type and content or content_url
- Artifacts for file-like outputs (e.g., images, JSON)
- Agent Manifest metadata for discovery
- Client SDK interactions (Python) for run management and streaming
Caveats
- The design emphasizes REST-based endpoints and SSE streaming; it does not rely on JSON-RPC.
- Supports asynchronous-first workflows with optional synchronous execution and streaming updates via SSE.
- License is MIT as per repository metadata.
# Agent Communication Protocol (ACP) ACP is an open protocol for agent interoperability that solves the growing challenge of connecting AI agents, applications, and humans. It enables standardized communication across different frameworks and infrastructures. ## Key Features - **REST-Based Design** - Straightforward HTTP endpoints (not JSON-RPC) - **Multi-Modal** - Supports text, images, audio, video, and any MIME type - **Async-First** - Built primarily for asynchronous communication with sync support - **Streaming** - Real-time Server-Sent Events (SSE) for incremental updates - **Stateful/Stateless** - Sessions for maintaining state across interactions - **Offline Discovery** - Agents remain discoverable through embedded metadata - **Framework Agnostic** - Works with BeeAI, LangGraph, CrewAI, LlamaIndex, and more ## Installation ```bash # Create new project with uv uv init --python '>=3.11' my_acp_project cd my_acp_project # Add ACP SDK uv add acp-sdk # Or with pip pip install acp-sdk ``` ## Quick Start: Creating an Agent ### Basic Echo Agent ```python import asyncio from collections.abc import AsyncGenerator from acp_sdk.models import Message from acp_sdk.server import Context,
- Key Features
- Installation
- Quick Start: Creating an Agent
- Basic Echo Agent
- Agent with Metadata
- Agent Manifest
- Required Fields
- Optional Metadata
- Message Structure
- Message Roles
- Message Parts
- Common MIME Types
- REST API Endpoints
- Discovery
Create new project with uv uv init --python '>=3.11' my_acp_project cd my_acp_project Add ACP SDK uv add acp-sdk Or with pip pip install acp-sdk uv run agent.py List all agents curl http://localhost:8000/agents
What does the agent-communication-protocol skill do?
Open protocol for AI agent interoperability enabling standardized communication between agents, applications, and humans across different frameworks
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill agent-communication-protocol-enuno-claude-command-and-c-2 --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.