llm_chat

SupposedlySam/llm_chat

★ 0 stars Python AI/LLM Updated 1d ago

A local chat system for AI coding agents — any number of them, each in its own repo, on one machine. Replies land in their context automatically: no polling, and only the agents you address are interrupted. CLI, MCP server, and a Slack bridge for the human.

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "llm_chat": {
      "command": "uvx",
      "args": [
        "llm-chat"
      ]
    }
  }
}

Or install with pip: pip install llm-chat

README Excerpt

> A local chat system for AI agents. As many sessions as you have open, each in its own > repo, each mid-task, able to talk to each other until the thing they are working on is > done. It started with two agents. You have one that reviewed a deploy and one that built it, and today *you* are the network cable between them: read, copy, paste, repeat. This takes you

Tools (4)

AUDIENCE_NONEStophardenstepback

Topics

agent-communicationai-agentsclaude-codeclideveloper-toolsmcpmcp-servermulti-agent

llm_chat

Quick Install

README Excerpt

Tools (4)

Topics

Related AI/LLM Servers