How Testnova fits together
Testnova turns plain-English intent into running tests, and running tests back into a report you can act on. Everything flows through one pipeline: you describe what your app should do, Testnova writes the test cases, generates runnable scripts, executes them on real devices, and sends you the result.
Each stage is its own object you can open, edit, and re-run, so you’re never staring at a black box. You can read the generated case, inspect the script, watch the run live, and trace any failure back to the exact step that broke.
The objects you’ll work with
- Project, your app under test: its URL, a login, and the real screens Testnova captures during setup.
- Test case, one feature described in plain English, e.g. “a user can reset their password.”
- Test script, the runnable automation Testnova generates from a case, saved to your own repo.
- Run, a single execution of a script (or a whole suite) on web, a real phone, or your API.
- Report, the pass/fail summary, step screenshots, and triaged bugs delivered after a run.
- Bug, a confirmed failure, separated from flaky noise, with the steps needed to reproduce it.
Grounding, why the tests actually work
Most AI test tools guess at your buttons and break on the first real click. Testnova grounds itself in your real product instead. During setup it signs in, captures your actual screens, and records the real selectors, links, and form fields it sees. Every script is built from what’s genuinely there, not from a hallucinated guess.
When your app changes and an element moves, Testnova re-grounds against the live screen, repairs the script, and opens a fix. Your suite keeps passing instead of quietly rotting between releases.
The REST API
Anything you can do in the dashboard, you can do over the API. Authenticate with a project key as a bearer token in the request header: Authorization: Bearer tn_live_…, create and revoke keys under Settings → API keys.
Keys are scoped to a single project and can be rotated at any time without touching your code.
- Projects, create a project, list projects, and fetch a project’s status.
- Cases & scripts, generate cases from a requirement, turn cases into scripts, and list or fetch either.
- Runs, start a run, poll its status, and pull the finished report and any bugs.
- Webhooks, get notified when a run completes so your own systems can react.
Integrations
- GitHub & GitLab, generated scripts and self-heal fixes land as commits or pull requests in your own repo.
- Issue trackers, push confirmed bugs to Jira and similar tools, complete with steps to reproduce.
- Notifications, run summaries and regression alerts go to email and Slack.
- CI, trigger a run from your pipeline and gate a release on the result.
Security & data handling
- Your test scripts live in your repository, not locked inside Testnova.
- The logins Testnova uses to reach your app are encrypted per project and never written into scripts or reports.
- Runs execute in isolated cloud environments that are torn down after each job.
- Revoke an API key or disconnect a repo at any time, and access stops immediately.