Command
Send feedback or report a problem about the qa plugin — opens a pre-filled GitHub issue (plugin version, OS, and what you were doing auto-filled) so the maintainer can collect & fix it. No login needed to view the link; uses gh to file it directly if installed.
Command
Extract durable locators for a screen — auto-route by platform (web Playwright / android / ios), reading only the matching platform skill, nothing extra.
Command
Fix a bug (compile/test fail, flaky, rule violation) at the correct layer, or update an existing plan. Auto-route by platform to get locators/read the correct platform rules.
Command
Generate MANUAL TEST CASES from a requirement/spec/Figma or a test-case plan → xlsx + (preferably) Google/Lark Sheet; parallel multi-phase when large.
Command
Introduce & guide usage of the qa plugin — which commands, which skills, the automation vs manual workflows, and the platform router.
Command
Kill all running Appium servers using the kill-appium.sh script.
Command
Review → fix → integrate latest target via a temp branch (safe merge) → push → open a Pull/Merge Request (auto-detects GitHub gh / GitLab glab from the remote).
Command
Manage "test-id debt" — EXPORT the list of elements missing an id to send to dev, RECORD a new element, or RESOLVE once dev has added the id.
Command
Build a multi-phase MANUAL TEST-CASE plan (scope, screens, phases, TCID range) → plans/ / .md + per-phase files, detailed enough for /qa:gen-testcases to run.
Command
Exploratory gate first, then design an AUTOMATION test plan (plans/ /plan.md). By DEFAULT, plan even when bugs are found (record them as known-issues/risks) as long as elements are declarable + actionable; pass --stop-on-bug to STOP at a bug report instead. Auto-routes by platform to read the correct design-pattern…
Command
Produce a blameless incident postmortem — timeline, root-cause analysis (5 Whys + technical/process contributing factors), impact, resolution/recovery, what-went-well/wrong/lucky, and a tracked action-item table (Prevent/Detect/Mitigate, owner role, priority, due). Grounds the timeline in optional related bug records…
Command
Review code (review-audit) → auto-fix clear Blockers → green build → commit & push the current branch to origin.
Command
Aggregate QA quality metrics into a manager-facing dashboard report — pass rate + trend, open bugs by priority with aging, defect density / hot-spot modules, created-vs-resolved trend, coverage per feature. Reads test runs from results/tests/, the bug register, and live Lark Bitable records (read-only). Output md (+…
Command
Score a release GO / NO-GO / CONDITIONAL against an auditable checklist — open bugs by priority + aging, pass rate, coverage vs the team's gate criteria in release-gate.config.yml. Hard gate fail => NO-GO; soft gate fail => CONDITIONAL (ship with sign-off). Reads the bug board + results/ (read-only), writes…
Command
Generate release notes for TWO audiences from a release — an internal technical changelog (Conventional-Commit grouped, with fixed-bug table) + a user/stakeholder-facing notes doc (plain language, benefit-led, grouped by highlight). Reads git commits since the previous tag + fixed bugs from the Lark board (read-only).…
Command
Review the current git change (diff) against the design system & coding rules, output findings with code + how to fix.
Command
Review the ENTIRE codebase against the design system, coding rules and cross-file consistency.
Command
Assess quality risk for a feature/release from a QA-Lead perspective — a Risk Matrix (Likelihood × Impact, 1–25) classified Low/Medium/High/Critical, with prevention/detection/response/owner mitigation for every Medium+ risk and a risk-based test strategy (coverage focus, exploratory areas, regression scope, staged…
Command
Compile then run tests + summarize the report; auto-route by platform (web Playwright / app Appium), reading only the matching platform skill.
Command
Scaffold a brand-new standard QA automation project from scratch (Playwright-Web or Appium-App, Page Object Model). Lays the canonical folder structure + static boilerplate (pom/Makefile/suites/configs/.env/.gitignore/sitemap), generates the framework + ONE example feature (auth/login + GoToHome), and verifies it…
Command
One-click setup of the qa plugin in THIS project — you (Claude) auto-detect the OS and run scripts/setup.py yourself; it creates .claude/qa-claude/ (config + .plugin.env + README usage guide), patches .gitignore, and checks the toolchain. The user does NOT run any terminal command.
Command
Discovery crawl — walk the whole app/web target (or just one feature), declare every screen's elements (name + stable locator only) and build/refresh the navigation sitemap under sitemap/. Pure mapping — NO tests, NO bug hunting, NO Page Object code. Auto-route by platform, reading only the matching skill.
Command
Evaluate SLA compliance from ticket/bug data and produce a period report (sprint/month/quarter) — compliance rate (overall + per priority), MTTR-Response & MTTR-Resolution with P50/P90/P95 percentiles, breach analysis, trend vs previous period, assignee performance. Reads a sheet/file/pasted data or the Lark board…
Command
Quick overview of the project state (git, devices, Appium, coverage).