Skip to main content

Documentation Index

Fetch the complete documentation index at: https://quashbugs.com/docs/llms.txt

Use this file to discover all available pages before exploring further.

When you run a task for the first time, Mahoraga operates in full reasoning mode - reading each screen, identifying elements, handling interruptions, and making decisions at every step. When the run completes successfully, Mahoraga stores the full path it took. From the second run onwards, Mahoraga follows that stored path directly. For each step, it checks the current screen against what it expects. If the screen matches, it executes immediately — no inference, no reasoning pause. A test that took around 5 minutes on the first run typically completes in around 1 minute on reruns. No configuration is needed. If you have already run a task successfully, the next run will automatically use the stored path.

​
How Mahoraga handles UI changes

Mahoraga does not fail when it encounters something unexpected during a rerun. Instead, it switches back into reasoning mode for that specific step:
  • A button has been renamed — Mahoraga identifies it by role and context, not just its label.
  • A permission dialog appears mid-flow — Mahoraga handles it and returns to the main flow.
  • A promotional modal blocks the screen — Mahoraga dismisses it and continues.
  • The layout has shifted on a different device — Mahoraga finds the correct element and continues.
Once it resolves the change, Mahoraga updates the stored path. The next rerun already knows what to expect.
⚠️ Note: Rerun speed gains apply after the first successful run. If the first run did not complete, no path is stored and the next run starts fresh.