Launching today
testcat
AI agents test your iOS app while you watch live
4 followers
AI agents test your iOS app while you watch live
4 followers
Open-source macOS app: describe a test in plain language, an AI agent drives real iOS simulators (or your iPhone) and hands back an evidence-backed verdict — while you watch live.
How does it actually handle flaky tests or simulator states between runs, and is there any way to step in manually if the AI agent starts going off-script mid-test?
@sibelkdrw4ar Good question — this is exactly what agent profiles are for. Each profile has an editable system prompt and a skills list, so you can give the agent explicit directives: recovery rules for flaky states, hard boundaries ("never touch account settings"), or step-by-step house procedures via a custom skill — and it loads them on every run. You can also stop a run at any moment while watching its reasoning live. More advanced mid-run controls (like steering a running agent) are planned.
Thanks for asking!