Skill Detail

Coordinate Terminal Coding Agents With hcom

Use hcom to launch, message, watch, fork, and coordinate multiple terminal coding agents across Claude Code, Codex, Gemini, OpenCode, Cursor, and similar CLIs.

Developer ToolsMulti-Framework
Developer Tools Multi-Framework Security Reviewed
โญ 393 GitHub stars
COPY SKILL INSTRUCTIONS (OPTIONAL)
npx skills add agentskillexchange/skills --skill coordinate-terminal-coding-agents-with-hcom Copy
Uses the third-party skills CLI, not an ASE-owned installer. Check your agent’s compatibility. This copies instructions; complete the upstream tool setup below separately.
At a glance
Tools required
hcom CLI installed by Homebrew, release installer, uv, or pip; terminal-agent CLIs such as Claude Code, Codex, Gemini, OpenCode, Cursor Agent, Kimi, or Copilot
Install & setup
uv tool install hcom
Author
aannoo
Publisher
Individual
Last updated
Jul 21, 2026
Quick brief

hcom is a terminal coordination workflow for teams of local coding agents. The operator starts agent CLIs through hcom, gives each agent an identity, then uses hcom messaging, event watching, transcripts, terminal views, spawning, forking, and collision notifications to coordinate work across terminals.

How it works

What this skill actually does

Use this instead of manually copy-pasting between agents when multiple CLI agents need to split work, review each other’s changes, report status, or hand off context. It is useful for parallel investigation, cross-agent review, headless subagents, and long-running sessions where the operator needs a lightweight control plane without changing the underlying agent tools.

The scope boundary is peer-agent terminal coordination through hcom’s local CLI, hooks, and SQLite-backed message/event store. It is not a generic multi-agent framework, broker, or chat app. The approved use is launching and supervising concrete Claude Code, Codex, Gemini, OpenCode, Cursor, Kimi, Copilot, or similar terminal-agent sessions while preserving their normal command-line behavior and ergonomics.