{"slug": "show-hn-async-multi-agent-coordination-across-physical-environments", "title": "Show HN: Async multi-agent coordination across physical environments", "summary": "A developer released Agent Comms, an open-source protocol and toolkit that lets AI coding agents share context and coordinate tasks across sessions and machines, installable via pip from the GitHub repository BlahBlah23406/agent-comms. Agent Comms packages an agent's task roadmap, architectural decisions, rejected hypotheses, and uncommitted git diffs into portable \"Context Capsules\" for out-of-session handoff, and adds an AHRP Live Relay WebSocket mesh for peer discovery, publish/subscribe messaging, and cross-machine Remote Procedure Calls. The toolkit ships an MCP server exposing export_handoff_capsule, import_handoff_capsule, and list_saved_capsules, and works natively with Claude Desktop, Antigravity, and Cursor via the Model Context Protocol.", "body_md": "A lightweight protocol and toolkit for AI coding agents to share context, hand off tasks across sessions, and collaborate in real-time across machines.\n\nWhen working with AI coding agents (Claude Desktop, Antigravity, Cursor, etc.), two major challenges arise:\n\n1. **Context Loss Across Sessions:** Starting a new chat or moving to another machine resets the agent's mental model and working memory.\n2. **Multi-Agent Coordination:** Agents running in different environments or physical computers cannot easily communicate, exchange state, or run coordinated tasks.\n\n**Agent Comms** provides a unified solution:\n\n- **Context Capsules (Out-of-Session Handoff):** Packages an agent's task roadmap, architectural decisions, rejected hypotheses, and uncommitted git diffs into a compact, portable bundle. The next agent session resumes immediately without burning context tokens.\n- **AHRP Live Relay (In-Session Collaboration):** A real-time WebSocket mesh supporting peer discovery, publish/subscribe messaging, and cross-machine Remote Procedure Calls (RPC).\n- **Native MCP Integration:** Works out-of-the-box with Claude Desktop, Antigravity, and Cursor via the Model Context Protocol.\n\nAgent Comms captures cognitive state alongside your working tree without polluting Git commit history:\n\n- **State & Decisions:** Records what worked, what was rejected, and the immediate next steps.\n- **Code Diffs:** Captures staged, unstaged, and untracked changes into a clean patch.\n- **Briefing Generation:** Produces a token-efficient Markdown briefing tailored for the incoming agent.\n\n```\nMachine A (Active Session)                  Machine B (New Session)\n ┌──────────────────────────┐                ┌──────────────────────────┐\n │ Agent exports capsule    │──[File/Sync]──▶│ Agent imports capsule    │\n │ (diffs + state + roadmap)│                │ (restores diffs + state) │\n └──────────────────────────┘                └──────────────────────────┘\n```\n\nFor multi-agent workflows, a lightweight relay server coordinates agents over WebSockets:\n\n- **Peer Discovery:** Agents announce presence, roles, and hardware capabilities.\n- **Cognitive Blackboard:** Replicated state where agents share real-time decisions and learnings.\n- **Direct RPC:** Agents can invoke tools or run commands on peer machines.\n\n**Using Pip:**\n\n```\npip install git+https://github.com/BlahBlah23406/agent-comms.git\n```\n\n**Or Clone & Install Locally:**\n\n```\ngit clone https://github.com/BlahBlah23406/agent-comms.git\ncd agent-comms\npip install -e .\n```\n\n**One-Line Install Script:**\n\n- **macOS / Linux:**\n\n```\ncurl -sSL https://raw.githubusercontent.com/BlahBlah23406/agent-comms/master/install.sh | bash\n```\n\n- **Windows (PowerShell):**\n\n```\nirm https://raw.githubusercontent.com/BlahBlah23406/agent-comms/master/install.ps1 | iex\n```\n\nRun initial setup:\n\n```\nagent-comms setup\n```\n\nBefore ending a session or switching computers:\n\n```\nagent-comms capsule pack \\\n  --task \"AUTH-01\" \\\n  --summary \"Migrated auth module to JWT; integration test pending\" \\\n  --next \"Run pytest tests/test_auth.py\" \\\n  --learning \"finding:PyJWT requires algorithms=['HS256']\"\n```\n\nRestore your uncommitted files and task briefing:\n\n```\nagent-comms capsule unpack \"AUTH-01\"\n```\n\nSee two local agents discover each other and collaborate:\n\n```\nagent-comms demo\n```\n\nAgent Comms includes an MCP server exposing `export_handoff_capsule`, `import_handoff_capsule`, and `list_saved_capsules`.\n\nAdd to your MCP settings file:\n\n```\n{\n  \"mcpServers\": {\n    \"agent-comms\": {\n      \"command\": \"agent-comms\",\n      \"args\": [\"mcp\"]\n    }\n  }\n}\n```\n\nOnce added, interact naturally with your agent:\n\n*\"Save my progress into a handoff capsule for task AUTH-01.\"*\n\n*\"Resume task AUTH-01 from my latest capsule.\"*\n\n- [**Natural Language User Guide** (`USER_GUIDE.md`)](https://github.com/BlahBlah23406/agent-comms/blob/master/USER_GUIDE.md) — Plain-English prompt examples for Claude Desktop, Antigravity, and Cursor.\n- [**Onboarding Guide** (`ONBOARDING.md`)](https://github.com/BlahBlah23406/agent-comms/blob/master/ONBOARDING.md) — Step-by-step developer onboarding and distributed agent recipes.\n- [**Protocol Specification** (`SPECIFICATION.md`)](https://github.com/BlahBlah23406/agent-comms/blob/master/SPECIFICATION.md) — Formal AHRP wire protocol and JSON schemas.\n- [**Tutorial & Recipes** (`TUTORIAL.md`)](https://github.com/BlahBlah23406/agent-comms/blob/master/TUTORIAL.md) — Hands-on walkthroughs and implementation examples.", "url": "https://wpnews.pro/news/show-hn-async-multi-agent-coordination-across-physical-environments", "canonical_source": "https://github.com/BlahBlah23406/agent-comms", "published_at": "2026-09-15 15:40:45+00:00", "updated_at": "2026-09-15 15:50:25.028203+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "developer-tools", "ai-products"], "entities": ["Agent Comms", "Claude Desktop", "Antigravity", "Cursor", "Model Context Protocol", "BlahBlah23406", "AHRP Live Relay", "Context Capsules"], "alternates": {"html": "https://wpnews.pro/news/show-hn-async-multi-agent-coordination-across-physical-environments", "markdown": "https://wpnews.pro/news/show-hn-async-multi-agent-coordination-across-physical-environments.md", "text": "https://wpnews.pro/news/show-hn-async-multi-agent-coordination-across-physical-environments.txt", "jsonld": "https://wpnews.pro/news/show-hn-async-multi-agent-coordination-across-physical-environments.jsonld"}}