Surfaces
Start a task by voice, continue in Desktop or the CLI
You describe a feature by voice on your phone, Devin opens a draft PR, and you continue the same session in Devin Desktop or the CLI.
Steps
- Enter your GitHub user or org. The links and the prompt on this page use it.
- Create your own copy of the starter repo. Open GitHub's new-repository page for devin-demo-voice-to-desktop, choose your account as the owner, keep the name
devin-demo-voice-to-desktop, and click Create repository. - Make sure Devin can access the new repo under Settings → Connections → GitHub.
- Install Devin Desktop, or the CLI:
curl -fsSL https://cli.devin.ai/install.sh | bash. - On your phone, open Devin, start a voice call and say the prompt below.
Prompt
Hey Devin — on https://github.com/<you>/devin-demo-voice-to-desktop, add a 90-day uptime bar to each component on the status page, like the ones on GitHub's status page: green/amber/red per day computed from incident history, with a hover tooltip. Get it working and push a draft PR; I'll pick it up from my laptop to review and tweak.
When the draft PR is up, stop and summarize where you left off and exactly what's left, so this can be continued from Devin Desktop/CLI with full context.What to watch
- The call creates a normal session that you can open from any device.
- Devin stops at a draft PR with a summary of what's left.
You're done when
- A draft PR is up.
- You picked the session up on your laptop, steered it and merged.