qlaneai/qlane
AI QA that runs your app in a browser on every pull request: set up a test target, read what it found on a pull request, and plan tests for a local diff. Ships one server per data region — connect the one hosting your organization.
Show what Qlane found on a pull request or a recent test run — defects, failing and blocked cases, the steps to reproduce them and their evidence — then locate the likely cause in this repository and propose a fix. Read-only on Qlane, and edits nothing without a yes. Use when the user asks what Qlane found, about Qlane results or bugs, about a failed Qlane check on a pull request, or what to fix — including plain requests in a repository connected to Qlane that do not name it, such as "check the results of the test", "what did the tests find", "what did qlane find", "show me the bugs from the last run" or "why did the tests fail on my PR".
Set up Qlane testing for this repository — find or create its Qlane project, create or fix a test target, have the developer enter its variables and test credentials on a Qlane page, run a configuration check and act on its diagnosis, then recommend switching on pull-request testing. Use when the user asks to set up, connect or onboard a repository to Qlane, to make Qlane's pull-request testing work, or when a configuration check failed — including plain requests such as "set up qlane for this repo", "connect this repo to qlane", "make PR testing work", "get qlane testing my pull requests" or "why did the configuration check fail". Asks before every change to the repository or to Qlane's configuration.
Test uncommitted local changes before opening a pull request. Asks Qlane to plan what should be tested for the current diff, then exercises those cases against the running app — driving the browser, reading logs and checking data where available — and reports suspected bugs for the developer to triage. Use when the user has local edits and asks to test them, check for regressions, verify a change before pushing, or find what their change might have broken — including plain requests such as "test my changes", "test this before I push" or "did I break anything".