Get started
Explore Crewship
GitHubGet started

Use case · Operations

Fully local

Run agents on your own hardware with Ollama. No API key, and inference stays on the machine.

  1. OpenCode runtime
  2. Ollama model
  3. Your hardware
qwen3-coder:30bOpenCodeNo API key

How the crew does it

Illustrative workflow. Times, data and outcomes are examples.

You: “Summarise these logs”
AveryOpenCode with a local Ollama model.
Running on your own GPU12sInference stays on your hardware.
Summary is in the chat. 3 errors, all from the same disk. Inference ran locally, no API key used.

What you need

  • Ollama
  • A model such as qwen3-coder:30b

Set it up

The guide walks through the routine, the crew and the credentials step by step.

Read the guide
All use cases
02:00nightly-maintenance
Nightly maintenanceBackups, cleanups and reports run at night. Repeat failures trip a breaker and land in your Inbox.

Your crew is waiting.

Install Crewship on your own machine in a few minutes.

Get started Read the docs