skills.rest
cbd
cbdCOMMUNITY
Explore all skills by this author →
𝕏@chaddepueGitHub

clawchat

Coordinate AI agents via a local chat server using TCP and Unix sockets.

1|Updated Mar 1, 2026
Search Tags:#code review#task delegation#decision-making#ai coordination#local chat server
One-click install
npx skills add https://github.com/cbd/clawchat --skill clawchat
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: clawchat
Source: https://github.com/cbd/clawchat/tree/main/skill
Command: npx skills add https://github.com/cbd/clawchat --skill clawchat

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires clawchat, and includes scripts (resource) and references (resource) components.

What problem does it solve?

ClawChat solves the challenge of coordinating multiple AI agents working on the same codebase by providing a local chat server for real-time communication, decision-making, and task delegation.

Core Features & Use Cases

  • Real-time Communication: Agents can send messages and notifications to each other in rooms.
  • Sealed-ballot Voting: Facilitates group decisions without anchoring bias.
  • Leader Election: Allows one agent to break ties and make binding decisions.
  • Message History: Persistent record of communication for accountability.
  • Use Case: Imagine coordinating a team of AI agents for code review. You can create a room, discuss issues, vote on pull requests, and elect a leader to make final decisions.

Quick Start

Start the ClawChat server and send a message to the 'code-review' room: clawchat send code-review "Review PR #123 - please check the code."

Frequently Asked Questions about clawchat

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I coordinate multiple AI agents for code review on the same machine?▼

To coordinate AI agents for code review, use a local chat server that enables real-time communication, task delegation, and decision-making via TCP and Unix sockets. Agents send messages to specific rooms to discuss issues and review pull requests.

How does sealed-ballot voting work for AI agent decision-making?▼

Sealed-ballot voting for AI decision-making allows agents to submit votes without seeing others' choices first, preventing anchoring bias. A leader agent is elected to break ties and make binding final decisions.

Do I need the clawchat binary and auth.key to run a local AI chat server?▼

Yes, running a local AI chat server requires the clawchat binary and a valid auth.key file. These components handle the TCP and Unix socket communication protocols that agents use to connect and exchange messages.

What's the best way to delegate tasks among AI agents working on the same codebase?▼

Task delegation among AI agents on the same codebase is best handled through a local chat server with persistent message history for accountability. Agents communicate in rooms, assign tasks, and vote on decisions for coordinated execution.

Can I use local chat server communication for AI agents over TCP and Unix sockets?▼

Local chat server communication for AI agents utilizes both TCP and Unix sockets for message transmission. This enables reliable real-time communication and notification delivery between agents operating on the same machine.

When do I need leader election for AI agent coordination?▼

Leader election for AI agent coordination is needed when group decisions result in ties or require a binding final call. One agent is elected as leader to break deadlocks, ensuring code review and task delegation processes are not stalled.

Related Skills

Explore More Software Engineering Skills→
project-axproject-ax
5

cli

Manage AX agents via command-line chat, messaging, and plugin operations.

Official
Intermediate
curiositechcuriositech
2

ongaro-ousterhout-2014-raft

Design distributed consensus protocols using Raft's decomposition patterns and append-only log constraints.

Official
Advanced
zhenzhizhi12zhenzhizhi12
2

cangjie-socket

Implement TCP, UDP, and Unix Domain Socket clients and servers in Cangjie.

Community
Intermediate
zenobi-uszenobi-us
68

mcp-developer

Develop MCP servers and clients with JSON-RPC 2.0 and schema validation.

Official
Advanced