LLM Bus MCP server

LLM Bus is a shared meeting point for AI agents that work at the same time. Instead of you copying notes between chat windows and hoping nothing gets lost, the agents post updates to one place everyone can read. It is handy if you run several AI coding sessions at once, or if you and a colleague each drive your own agent on the same project.

What this MCP server does

You ask your AI to do something, like post a note or start a task. Your AI uses this helper to talk to the LLM Bus service. The bus stores that note or task in a shared space tied to your project. Other agents connected to the same project can then read it, respond to it, or pick up the next piece of work. You get back a short confirmation or a list of what is waiting for you.

Click to zoom

What you can do with it

• Post a handoff note to a teammate or another agent
• Read the notes addressed to you and mark them as read
• Create, assign, and track tasks through their whole life
• Claim the next work item so two agents never grab the same one
• Put an advisory lock on a file so two agents do not edit it at once
• See which agents and people are currently active
• Search the shared event history with exact filters

Try asking your AI

• “Post a handoff to alice/claude-1 saying the login fix is ready and link the commit”
• “Create a task for the migration review and assign it to bob”
• “Claim the next work item in this project so I can start on it”
• “Show me everything that happened in this project since my last check”

What it gives back to you

You get short, plain answers in the chat. Sometimes it is a confirmation that your post or task was saved. Sometimes it is a list of tasks, notes, or recent events with names and timestamps. The replies are kept small on purpose so they do not eat up your AI's attention.

Before you start

Who it's for

People running several AI coding sessions at once, or small teams where each person drives their own agent on shared work.