How to Deploy KiroCrew with Docker and Run Unattended Dev Tasks
Step-by-step tutorial: run KiroCrew with Docker, connect the CLI or web dashboard, and set up scheduled unattended tasks and heartbeats.
🚀 Want to deploy KiroCrew yourself?
Docker configs, system requirements, and installation guides — all on one page.
View KiroCrew Tool Page →KiroCrew is a persistent AI development workspace — you give it a multi-step task, and it keeps going even after you disconnect. In this tutorial we'll get it running with Docker, connect from the CLI, and set up a scheduled job.
Step 1 — Run with Docker
The official image is published on GitHub Container Registry:
docker run -d --name kirocrew \
-p 8080:8080 \
-v ./data/kirocrew:/data \
ghcr.io/kirodotdev/kirocrew:latestStep 2 — Connect a client
Once the container is up, open the web dashboard at http://localhost:8080 or use the CLI. The desktop app (macOS/Linux) and Slack/Discord connectors all attach to the same persistent workspace, so you can switch surfaces mid-task without losing state.
Step 3 — Run your first unattended task
- Create a new project workspace from the dashboard.
- Describe a multi-step task — e.g. "refactor the auth module, add tests, and open a PR".
- Start the task and close the terminal. KiroCrew keeps executing.
- Check back via the dashboard or a Slack heartbeat when it finishes.
Pro tip: use recurring schedules for nightly builds or dependency updates, and heartbeats to monitor services — KiroCrew only pings you when something needs attention.
System requirements
| Tier | CPU | RAM |
|---|---|---|
| Minimum | 2 cores | 4 GB |
| Recommended | 4 cores | 8 GB |
KiroCrew is Apache-2.0 licensed with an active community (2.3k+ stars). It's a strong pick if you want coding agents that run unattended on infrastructure you own.
🚀 Want to deploy KiroCrew yourself?
Docker configs, system requirements, and installation guides — all on one page.
View KiroCrew Tool Page →