Test a changed user interaction for realistic sequence failures such as double submission, stale responses, navigation, and recovery; use for interaction QA rather than general code review.
The English summary is being prepared. Showing the original description.
Run score Not tested
See how to installReview the instructions the agent will receive before you install.
And what happens if I click it twice?
Read supplied requirements, interaction code, native tests, existing test support and applicable project instructions together when practical. Discover missing paths, not another inventory before known-file reads. Use controlled responses/clocks, isolated cases, bounded waits, and owned-operation cleanup—never sleeps or production actions. Preserve user files and scope.
Derive expected states from the product contract, not today's implementation. Separate final ownership from interval requirements. “Retain while loading” or “until the latest completes” applies throughout that interval: assert the retained value after relevant intervening completions while the latest remains pending, not only at request entry. Printing an observation is not a regression assertion. If only final ownership is required, leave unspecified intermediate display as observation; seeing an older response does not require a valid fix to display it. A regression must reject the fault without rejecting another permitted path.
Choose the requested deliverable; load only its relevant support:
Project regression: extend and run the native test directly, not a disposable probe first. Reuse adequate project fixtures and covered cases; skip transport and probe references when that support is already present. Only when additional Python request-control support is needed, read the native-test interface and copiable transport together. An installed-helper command or saved JSON is not a standalone project regression.
Captured observations without a project-test deliverable: the optional
component probe fits a zero-argument constructor,
async run(query, fetch) and direct payload state. An opt-in mode checks required
display retention through normal/reversed overlap; repeated identical overlapping
queries still require project checks. Existing adequate runners take precedence.
Never change production to fit a helper. Native-project delivery does not need
this reference.
Rendered UI: read references/browser.md.
Return the tested sequence, expected/observed state, tested layer, and complete bounded command. Stop after the nearest normal and targeted adversarial cases. Do not add separate evidence files when captured check output already proves them. Browser unavailability leaves browser QA incomplete. QA does not authorize a fix.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Audit whether tests detect broken behavior by tracing assertions and running narrow controlled mutations in an isolated workspace; use for test-quality reviews, not routine test execution.
The English summary is being prepared. Showing the original description.
Diagnose an uncertain bug by designing experiments that distinguish competing causes, especially speculative cache, timing, environment, or concurrency explanations.
The English summary is being prepared. Showing the original description.
Review a planned deployment or release diff for rollback feasibility, mixed-version compatibility, migrations, and configuration readiness without deploying it.
The English summary is being prepared. Showing the original description.
Keep a small requested change focused when optional refactors, architectural redesign, or unrelated cleanup threaten to expand its scope.
The English summary is being prepared. Showing the original description.
Review proposed abstractions, dependencies, and configuration for concrete consumers and maintenance cost when simplifying a design or checking overengineering.
The English summary is being prepared. Showing the original description.
Trace the purpose of suspicious legacy code using current callers, tests, and Git history when deciding whether it can be changed or removed.
The English summary is being prepared. Showing the original description.