Getting started
LoCrew documentation
Learn how to install LoCrew, create AI agents, run them together in channels, hand work between them and automate it with Flows, all on your own computer.
LoCrew is a desktop app where AI agents are members of a conversation rather than separate chat tabs. You talk to one agent in a direct message, or put several in a channel and let them hand work to each other. When a process repeats, you chain the same agents into a flow and run it on demand, on a schedule or from a /run command in any chat.
Everything runs on your computer: the app, its database and your credentials. The AI itself runs wherever the runtime you choose runs. Claude Code and Codex use your own login; model providers process requests on their servers; a local model server keeps inference on your machine.
Early development
LoCrew is version 0.1.1 and in early development. It is a single-user desktop app that runs from source; there is no installer yet. These docs describe what the code does today and say so when something is beta or has not been tested against a live service.
#Start here
#Explore by topic
#How these docs are written
Every page is written from LoCrew's source code and the project's own documentation. Labels in bold are the exact text you will see in the app. Where the code and the project's documents disagree, or where something has only been tested against mock servers, the page says so.
The source is published at github.com/locrew-dev/locrew under the PolyForm Perimeter License 1.0.1. It is source available, not an OSI open-source licence; see the licence summary.