A CLI that lets AI agents message, watch, and spawn each other across terminals.
hcom is easy to set up with trust notes worth reviewing. Check agent compatibility and use-case fit before adding it to your workflow.
Repository setup guidance
npx skills add aannoo/hcomRun the command in your terminal.
Confirm that the skill files were added to your agent workspace.
Check the README requirements before invoking the skill in your agent.
hcom is a command-line tool that connects different AI coding agents so they can talk to each other. You run it in front of your agent, and then agents can send messages, watch what others are doing, and even start new agents. It works with many popular AI tools like Claude Code, Codex, and Gemini.
hcom is a lightweight CLI tool that enables AI agents to communicate, observe, and orchestrate each other across terminals. It integrates seamlessly with popular AI coding assistants such as Claude Code, Codex, Gemini, OpenCode, Kilo Code, Pi, Antigravity, Cursor, Kimi, and Copilot without requiring changes to how you use them. Agents can message each other in real-time, observe transcripts and file edits, subscribe to events, and spawn, fork, resume, or kill each other. The tool uses a local SQLite database to record activity and deliver messages. It supports cross-device communication via an MQTT relay with end-to-end encryption. hcom is a single Rust binary with no background services, making it easy to install and use. It is ideal for coordinating pipelines, running different AI CLIs as subagents, or simply replacing copy-paste workflows.
Looks usable, but maintenance, license, or security notes deserve a closer look.
Last commit was about 51 days ago.
330 GitHub stars indicate community interest.
18 open issues signal maintenance load.
MIT license detected.
Coordinate a multi-agent pipeline where one agent reviews code and another fixes issues.
Spawn multiple agents to work on different parts of a problem in parallel and collect results.
Have an agent observe another's terminal to learn or debug without interrupting.
Fork an agent to investigate a bug while the main agent continues working.
Connect agents across different machines for remote collaboration.
Agents can execute arbitrary commands on your system; only use with trusted agents.
Cross-device relay uses end-to-end encryption, but treat join tokens like API keys.
Local SQLite database stores agent activity; ensure proper access controls.
330
Stars
42
Forks
18
Issues
MIT
License
Vercel's official collection of packaged instructions and scripts that extend AI coding agent capabilities.
Official Supabase agent skills to help AI agents work accurately with Supabase products.
A best-practices skill for Terraform and OpenTofu AI agents, enabling testing, module structuring, CI/CD, and production infrastructure code.
A CLI that lets AI agents message, watch, and spawn each other across terminals.
AI agent developers building multi-agent systems, Software engineers using multiple AI coding assistants, DevOps engineers automating complex workflows with AI
The setup section provides repository-level starting guidance, not a guarantee of an independently verified installation. Check the official README and release notes.
Agents can execute arbitrary commands on your system; only use with trusted agents. Cross-device relay uses end-to-end encryption, but treat join tokens like API keys.
Similar or complementary options to evaluate include agent-skills, supabase-agent-skills, terraform-skill.
Jun 6, 2026
3 security/trust notes recorded.
Setup difficulty is 2/5.