uiverify

Set up UI Verify visual regression testing in a repo end to end — detect or scaffold the capture path (Storybook stories, a Playwright suite, or Vitest browser-mode component tests), install the uiverify CLI and wire it into CI, and author the first stories the economical, deterministic way from the start. Use when adding visual/screenshot regression testing to a project for the first time, wiring UI Verify into CI, or onboarding a repo that has no visual tests yet.

Make @uiverify/vitest (Vitest browser-mode) captures deterministic so component visual tests stop coming back "changed" without a real change (flaky diffs). Use when setting up or debugging visual tests over Vitest browser-mode component tests. Focuses only on the run-to-run variation the capturer can't neutralize from outside your app — the clock, infinite JS animations, live data, and non-Math.random randomness — plus the one Vitest-specific trap: capturing before the component has settled.

Turn a code-owner correction (or a recurring mistake) into a durable rule - the right conventions-doc edit, and a real failing lint guard whenever the rule is mechanically checkable. This is the self-improving loop the factory leans on: when something goes rogue, capture it so the linter/conventions-doc catches it next time instead of a human re-explaining it. Also runs as a weekly sweep over recent merged work. Use when asked to "/add-rule", "turn this into a rule", "make sure this never happens again", "capture this correction", "add a lint guard for X", or "do the weekly rule review".

Raise a pull request from the current branch and babysit it to green. Opens the PR, then polls CI (~every 10 min) until every check passes, reacting to each failure by reading the logs, fixing locally, and pushing. When a failure exposes a generalizable lesson it proposes capturing it as a rule (via /add-rule). Does NOT approve or merge. Use when asked to "raise a PR", "open a PR and watch it", "push this and monitor CI", "get this PR green".

Start your dev stack on isolated ports, authenticate Playwright with the user's browser session, and test the current branch. Use when asked to "e2e verify", "playwright test", "test everything", "test my changes with playwright", or "test this branch".

Grade a completed factory run (a worktree + its chat) against your standards, cold and adversarially, so the system learns. Squashes the session transcript into a compact digest (prompts + corrections, the ordered tool timeline, the diff), then a fresh evaluator judges how the run went - did it follow the loop, honor your conventions, write real tests, ask at the right moments, handle corrections - and routes every finding to a durable fix (/add-rule for code, a persistent memory for process). This is the OFFLINE complement to the correction hook: the hook catches what you notice live, this catches what you didn't. Use for "evaluate this factory run", "grade this chat", "how did that go, what should it learn".

The umbrella "I say what I want, it ships it" loop. Clarifies the ask up front only if genuinely ambiguous, then implements it and drives it to a merge-ready PR - running the full local agentic loop (implement → review-loop → e2e-verify → review-loop → raise PR → monitor CI incl. visual → fix → re-review → converge) and stopping mid-way only when it truly needs a decision. Does NOT auto-merge. Use when asked to "ship this", "/factory", "build X end to end", "run the whole loop", "make it and get it green".

Iterative multi-model review-and-fix loop. Repeatedly runs the review-multi-model panel over the working tree, applies the verified fixes, and re-reviews - review → fix → review → fix - until a full pass finds zero verified issues (or a safety cap / no-progress guard stops it). Use when asked to "review-loop", "loop the review", "keep reviewing and fixing until clean", "review until no issues", "iterate review-multi-model until it's clean".

Multi-model code review. Runs the current branch (or a GitHub PR) through an independent Claude reviewer AND OpenAI Codex (ChatGPT) in parallel, verifies every finding against the actual code, and presents one merged, de-duplicated report - with an optional --fix step. Use when asked to "review-multi-model", "multi-model review", "panel review", "second opinion on this diff", "review with codex + claude".

Author economical visual-test stories — full visual coverage in the fewest billable snapshots. Applies to any per-snapshot visual tool (UI Verify, Chromatic, Percy, Playwright screenshots), which all bill and diff per rendered story. Use when a component's variants/states are exploding into one story per combination, when the visual-testing bill or noise is driven by snapshot count, or when writing new stories and you want the cheapest layout that still covers every variant. Covers gallery/matrix stories, data-driven states, and the coarse-granularity tradeoff.

Make @uiverify/playwright (archive-replay) captures deterministic so real-page visual tests stop coming back "changed" without a real change (flaky diffs). Use when driving a real page — your app under test, a staging build, or live production — and the diff flakes. A symptom-to-lever checklist for the run-to-run variation you probably have: feature flags, live data, the clock, third-party widgets, and dynamic layouts. The lever is almost always "stop the variation at record time" — the archive bakes in whatever the page was.

Make Storybook stories deterministic for UI Verify so captures stop coming back "changed" without a real change (flaky diffs). Use when setting up story-level visual tests or debugging a story that diffs every run. Focuses only on the run-to-run variation the capturer can't neutralize from outside your app — the clock, infinite JS animations, live data, and non-Math.random randomness — and deliberately skips what UI Verify already handles for you.

Triage a UI Verify build from your coding agent via the UI Verify MCP — bucket a build's changed stories into real regressions vs cosmetic reflow vs rendering noise, summarize the real regressions for a PR comment, and accept baselines in bulk. Use after a UI Verify build reports changes and you want the agent to review, summarize, or accept from the terminal instead of clicking through the dashboard. Triggers: "triage this build", "summarize the visual regressions", "accept all the new baselines".