From canary
Reads a git diff and generates a prioritized browser-QA plan of user-facing workflows affected by the change, optionally recording them as a session with a report.
How this skill is triggered — by the user, by Claude, or both
Slash command
/canary:canary-verifyThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Read a code change, infer the **user-facing workflows** it affects, and suggest a **prioritized QA
Read a code change, infer the user-facing workflows it affects, and suggest a prioritized QA
plan — the concrete Canary flows that verify them. Then optionally hand off to canary-session to
record those flows and produce report.html. Use the canary-scripting skill for the step API.
User says: "what should I QA for these changes?" or "verify my diff" Read the working-tree diff, map changed files to affected routes/flows, and present a ranked plan (P0/P1/P2). Offer to record the P0 flows as a session.
User says: "regression plan for feature/checkout" or "QA this PR" Diff the branch against its base, group the touched workflows, suggest the steps per flow, and hand the approved flows to canary-session for a report.
git diff and git diff --staged. A branch/PR:
git diff <base>...HEAD (list files with git diff --name-status <base>...HEAD). Or reason
straight from a prose description ("I changed the login redirect") — skip git.references/REFERENCE.md.references/REFERENCE.md.report.html; offer canary-review
to open it. Don't record flows the user didn't approve.session start / run / session end here.npx claudepluginhub 0xnyn/canary --plugin canaryDiff-aware AI browser testing that analyzes git changes, generates targeted test plans, and executes them via agent-browser. Use when testing UI changes, verifying PRs, or running regression checks on changed components.
Runs diff-scoped browser QA on the active branch: maps user flows, drives a real browser, autonomously fixes breakages with regression tests, and writes a dogfood report. Manual invocation only.
Analyzes repo tech stack and branch changes, traces impacts to user-facing entry points like CLIs and APIs, generates executable E2E YAML test plans.