Autonoma
AI agents that test your app end-to-end, no code needed
- Category
- Productivity & Automation
- Pricing
- Self-hosted: free (open source, BSL 1.1). Cloud pricing unverified - the official pricing page could not be fetched on 2026-10-27; check getautonoma.com/pricing
- Best for
- Good for small to mid-size engineering teams on GitHub who want PR-level end-to-end checks and have someone able to run Docker-based services.
- Official site
- getautonoma.com
- Last checked
- 2026-10-27
Autonoma suits engineering teams who want AI to run end-to-end browser tests on every pull request without hand-written Playwright scripts, and who are comfortable self-hosting or reviewing source code. Skip it if you need a fully priced, turnkey SaaS with published cloud rates, or if your testing is mostly mobile, since the repository describes web app testing. Per its GitHub README, AI agents drive a real browser against your app on each PR, and when a run fails Autonoma posts a review on the PR with expected versus actual behaviour, a screenshot or video, and the suspected code. The code is public under the Business Source License 1.1 (self-hosting and production use allowed, reselling as a competing service not allowed; it converts to Apache 2.0 on 2028-03-23). Self-hosting needs Node.js 24+, pnpm and Docker. We could not verify current cloud pricing, credits or customer claims on the official site today.
Good for small to mid-size engineering teams on GitHub who want PR-level end-to-end checks and have someone able to run Docker-based services. Not for teams that want a published, fixed cloud price before talking to anyone, non-GitHub workflows, or teams that need deterministic hand-written test suites and cannot tolerate agent-driven variability.
Key features
Agent-run browser tests on every PR
AI agents exercise your app in a real browser on each pull request, so you do not write test scripts or maintain selectors.
Failure reviews on the PR
When a run fails, a review is posted on the pull request with expected and actual behaviour, a screenshot or video, and the suspected code.
Self-hosting
The repository includes local setup steps (pnpm, Docker services, .env) so you can run the platform on your own infrastructure.
Source-available code
Code is public under BSL 1.1: read, modify, self-host and run in production, but no reselling as a competing commercial service.
Pricing breakdown
Self-hosted
- Public source under BSL 1.1
- Requires Node.js 24+, pnpm 10.x and Docker
- Community support via GitHub
Cloud / Enterprise
- Check getautonoma.com/pricing or contact the vendor for current plans and credit rates
Pros and cons
Pros
- Code is public, so you can inspect how the agent works and self-host instead of depending on a vendor.
- No test scripts or selectors to maintain: agents run the app in a real browser per PR.
- Failed runs come back on the PR with a screenshot or video and the suspected code, which shortens triage.
- Self-hosting is free of license fees, and the license converts to Apache 2.0 in March 2028.
Cons
- Cloud pricing could not be verified today, so budgeting is impossible from public data we could confirm.
- BSL 1.1 is source-available, not OSI open source until 2028; you cannot resell it as a competing service.
- Self-hosting needs Docker and recent Node.js, plus you carry hosting and AI model costs yourself.
- Agent-driven tests can behave less deterministically than scripted ones, so expect some flaky or surprising runs.
- The repository is young (about 238 stars) and no independent review-site ratings were found.
- Domains are confusing: the old autonoma.app blocks automated fetches, and the product website is getautonoma.com; verify you are on the right one.
Frequently asked questions
Is Autonoma open source?
The code is public under the Business Source License 1.1. You may read, modify, self-host and run it in production, but not sell it as a competing service. It becomes Apache 2.0 on 2028-03-23.
How much does Autonoma cost?
Self-hosting has no license fee. We could not verify cloud pricing on 2026-10-27, so confirm current plans with the vendor before relying on any number you see elsewhere.
What do I need to self-host it?
Per the README: Node.js 24+, pnpm 10.x and Docker, then clone the repo, install, start the Docker services, configure .env and run pnpm dev.
How is it different from writing Playwright tests?
You describe nothing in code; agents drive the browser on each PR and report failures with media and suspected code. The trade-off is less determinism than scripted tests.
Does it work outside GitHub?
The documented flow posts reviews on GitHub pull requests. Support for other hosts was not verified.
Ready to try Autonoma?
Head to the official site to explore pricing and start a free trial where available.
Visit Autonoma →