Omi can build anything. Omi CLI brings that to your terminal: an AI coding agent that works inside your repo, where you can see and approve every change. We launched it on Product Hunt today, and we'd love feedback from developers.
Try it
npm install -g @builditomi/cli
cd your-project
omi
Node.js 18.17+ on macOS, Linux or Windows (WSL). There's a Free plan to try it.
What a task looks like
You describe the job in plain words, for example:
> add input validation to the signup form and make the tests pass
Omi reads the repo, shows a plan, and then:
- every edit appears as a diff you approve (yes / no / always)
- it runs your build, linter and tests with your permission, and
/testkeeps fixing the code until the tests pass -
/undoputs back every file the last task changed, even without git, and/rewindgoes back further
Models
Auto picks the model for each turn from 13 models (DeepSeek, Gemini, Gemma, gpt-oss, GLM), so everyday edits stay cheap and hard problems get a stronger model. You can also bring your own: omi -m ollama:<model> runs fully local with no sign-in, and openai:, anthropic:, gemini:, deepseek: or openrouter: use your own API key.
New in 0.5.0
- Subagents that edit in parallel, each in its own git worktree, merged back three-way
- Built-in browser testing: Omi opens your app in headless Chrome, clicks through it, and reports console errors and failed requests
-
Background agents (
omi bg "task") that keep working after you close the terminal -
Cloud agents (
omi cloud "task") that run on your repo's GitHub Actions and open a pull request
Claude Code's .claude/agents, commands and skills work as they are.
Pricing
One plan in US dollars with clear 5-hour and weekly limits: Free, Pro $20/month, Pro+ $60, and a $200 tier. Turns on your own API keys or local models never count against your limits.
We'd love your feedback
Which task should Omi CLI get better at next? Tell us in the comments here or on the Product Hunt page.
Docs and changelog: cli.builditomi.com
Top comments (0)