asap-protocol
Active·★ 16·Apache-2.0·Updated 2026-10-22
ASAP is a production-ready protocol for agent-to-agent communication and task coordination. It addresses challenges like connection complexity, state drift, and fragmentation in multi-agent systems. It provides stateful orchestration, MCP integration, and advanced security features.
#agent-communication#agentic-workflows#ai-agents#ai-infrastructure#mcp-servers#model-context-protocol#multi-agent-systems#protocol
01
Features
01Stateful orchestration with task state machine and snapshotting
02Schema-first design using Pydantic v2 and JSON Schema
03Async-native with asyncio and httpx support
04MCP integration for tool execution and coordination
05Security features including Bearer auth, OAuth2/JWT, Ed25519 signed manifests, and mTLS
02
Compatibility
Python
Python 3.13+
Verified via docs
LangChain
LangChain integration
Verified via docs
CrewAI
CrewAI integration
Verified via docs
LlamaIndex
LlamaIndex integration
Verified via docs
OpenClaw
OpenClaw integration
Verified via docs
03
Quick start
1
$ uv add asap-protocol
04
Use cases
↳Multi-agent orchestration with complex workflows
↳Stateful workflows requiring persistence and resumability
↳Production systems needing high-performance, type-safe agent communication
05
Alternatives
remnote-mcp-server★ 30
MCP server and CLI for connecting AI agents, OpenClaw/Hermes workflows, local automation, and coding harnesses to your RemNote knowledge base via the RemNote Automation Bridge.
mcp_agent_mail_rust★ 210
Rust MCP server for multi-agent coordination: 34 tools, Git-backed archive, SQLite indexing, advisory file locks, and an interactive TUI console
claude-plugins★ 97
28 plugins and MCP servers for Claude Code — TDD, multi-agent orchestration, iterative refinement, binary RE, structured decisions. Install any skill in one command.
octave-mcp★ 54
OCTAVE protocol - structured AI communication with 3-20x token reduction. MCP server with lenient-to-canonical pipeline and schema validation.
nodebench-ai★ 15
NodeBench MCP - 260-tool Model Context Protocol server with progressive discovery, agent-as-a-graph embeddings, research optimization, and adaptive web scraping. Monorepo: MCP server + Convex backend + React frontend.
Comments
Log in to leave a comment
No comments yet. Be the first!