Happy slice of mcp pizzaMCP.pizza
omnigent

omnigent

MCP.Pizza Chef: omnigent-ai

This tool gives you one home base for running several AI coding assistants — including Claude Code, Codex, and Cursor — side by side, so you can compare their work or have one check another's output. Your session follows you: start a task on your computer, then pick it up on your phone or in a browser. You can invite a teammate to watch an assistant work live, set spending limits, or require your approval before it takes a risky action. Setup involves a command-line install and an API key or subscription to a model provider such as OpenAI, Claude, or Gemini.

Coding
Communication

Use This MCP client To

Watch my AI coding agent work from my phone Have one AI agent double-check another agent's work Pick up a coding task on my phone where I left off Set spending limits so an agent can't run wild Require my approval before an AI agent changes risky files Let a teammate watch and chat with my AI agent live Run an AI coding agent in the cloud, no laptop needed

README

Omnigent

The open-source meta-harness for all your AI agents.

Omnigent is an open-source meta-harness that gives you a common orchestration layer over Claude Code, Codex, Cursor, OpenCode, Hermes, Pi, and the agents you write yourself: swap or combine harnesses without rewriting, enforce policies and sandboxing, and collaborate in real time from any device — terminal, browser, phone, or the native desktop app.

PyPI version License: Apache 2.0 Discord Status: alpha

omnigent.ai · ⬇️ Download the macOS desktop app

The Omnigent desktop app: starting a new session, with pinned and project-grouped sessions in the sidebar

Why Omnigent?

Omnigent lets you:

omnigent FAQ

Which AI coding assistants does this work with?
You can run Claude Code, Codex, Cursor, and several other coding assistants side by side in the same place, along with any custom agents you set up yourself.
Do I need an account or an API key to use this?
Yes. You'll need either an API key from a model provider — such as OpenAI, Claude, or Gemini — or an existing subscription to one of those services.
How hard is it to set up?
It takes some technical know-how. You install it from the command line on your computer, so it's best suited for people already comfortable using AI coding assistants.
Can I use this to keep an eye on my AI assistant from my phone?
Yes — a session you start on your computer stays in sync, so you can check in or keep working from your phone or a web browser.
Can I use this to have one AI agent double-check another agent's work?
Yes — you can run multiple agents in the same session and ask one to review or build on what another produced.
What stops an AI agent from doing something risky without me knowing?
You can set up rules that pause the agent and ask for your approval before it takes an action you'd want to review first.
Can a teammate watch or help with my AI agent's session?
Yes — you can share a session so someone else can watch it work live, chat with the agent, or take over for a bit.
Do I need to keep my own computer running the whole time?
No — you can launch agents in cloud sandboxes instead, so your laptop doesn't need to stay on.