Agentlas-OS

MCP.Pizza Chef: agentlas-ai

This tool lets you describe a task in plain language and have it turn into a ready-to-use helper agent, complete with its own instructions and abilities. You can also borrow pre-built specialist helpers from a shared public library instead of building your own from scratch. Agents you create are saved to a private account, so you can pull them up again later on a different computer. It works inside Claude Code, Cursor, Codex, Gemini CLI, and several other AI coding tools, and setup involves running an install command from a terminal window.

Coding

Use This MCP server To

Turn a plain-language request into a ready-to-use helper agent Borrow a pre-built specialist agent instead of building my own Save a custom agent so I can reuse it later Pull up my agents again on a different computer Combine several helper agents into one team for a project Keep my agent-building work private to my own account

README

Agentlas One launch film

Agentlas Agent Lab banner

Agentlas OS

Build it or borrow it. The agents you create stay yours.
Turn a plain-language request into a runnable agent or team, borrow specialists from the public Agentlas Hub, and keep the agents you create available through your private, owner-scoped Agent Cloud.
Run them through supported hosts for the LLMs you already use. Hephaestus is the open-source engine underneath.

We are Agent Trust. Your agent is not a program. It is an asset. — Agentlas —

An agent you create is not tied to one model workspace or computer. To retrieve it from Cloud elsewhere, install Agentlas OS on a supported host and sign in.

Latest release License: Apache-2.0 LLMs

English · Korean · 中文 · 日本語 · हिन्दी

Paste to Install

Paste this into the LLM you are using now, such as Claude Code, Codex, Gemini CLI, Antigravity, or Cursor:

Agentlas-OS FAQ

Which AI coding tools does this work with?
It works inside Claude Code, Cursor, Codex, Gemini CLI, and several other AI coding assistants, so the same custom agents follow you no matter which one you have open.
Do I need an account or a key to use this?
Yes. You need to sign in with an account to save and retrieve your agents, and you use the AI model account you already have with your coding assistant to actually run them.
How hard is it to set up?
It takes some technical comfort — you run an install command from a terminal window rather than clicking a single "add" button, so it suits someone used to following short command-line instructions.
Can I use this to build a custom AI helper without writing any code?
Yes — you describe what you want the helper to do in plain language, and the tool turns that description into a working agent for you.
Can I use this to reuse the same helper agent across different computers?
Yes — agents you save are kept in your own private account, so you can pull them up again after signing in on another machine.
Will other people be able to see the agents I build?
No. Agents you build stay private to your account unless you choose to share them, though you can also browse and borrow agents that other people have made public.
Do I need to know how to code to use this?
Not to describe or run an agent, but setting the tool up does involve running a command in a terminal, so basic comfort with that step is helpful.
Does this work with Claude, ChatGPT, or Gemini?
Yes — it is designed to work with whichever AI model you already use, including Claude, OpenAI's ChatGPT models, and Google's Gemini, among others.