web-testlisted

Verify UI behaviour in a real browser — derive the scope from a task's acceptance criteria and/or the uncommitted diff, resolve the target environment and its auth state, drive the page through a browser-automation MCP (Claude in Chrome), exercise the happy path plus a required set of edge cases, capture console/network/visual evidence, write `{ID} web-test-summary.md` (executed test plan + reproducible bug reports) into the task folder, and record a WEBTEST verdict in the task log. Used by the Tester (and standalone on work-in-progress). Interactive sessions only — a headless orchestrated run must not drive a browser (each call stalls it on a governance approval card). Mutating a shared environment is HIGH; production is CRITICAL.
bibennurbani/geekstack-flow · ★ 3 · AI & Automation · score 69

Install: claude install-skill bibennurbani/geekstack-flow

# Web test ## When to use this skill Invoke when UI behaviour has to be *seen* to be believed and the automated suite can't show it: - A task is `IN_TEST` and one or more acceptance criteria are **visual or interaction-level** ("the banner shows the schedule name", "the dialog closes and the row updates"). - The user wants a **smoke check of work in progress** — uncommitted frontend changes, before review. - The change is deployed to a **shared environment** (dev/UAT/staging) and the question is *"does it actually behave there?"*. - The user says *"web test ES-1234"*, *"check this in the browser"*, *"click through it on UAT"*. **Do not use this skill** for: static code review (`review-diff`), the scripted suites (`verify` — unit/integration/Cypress), authoring new Cypress specs (that's the Coder's job — propose them from what you found here), or anything on **production** beyond read-only observation with a recorded approval. **Do not use this skill in a headless orchestrated run** — see [Orchestrated runs](#orchestrated-runs) below. ## Preconditions 1. **An interactive session with a browser-automation MCP.** In Claude Code that is **Claude in Chrome**: invoke the `claude-in-chrome` skill first, then load the MCP tools in **one** `ToolSearch` call (`select:mcp__claude-in-chrome__tabs_context_mcp,mcp__claude-in-chrome__tabs_create_mcp,mcp__claude-in-chrome__navigate,mcp__claude-in-chrome__computer,mcp__claude-in-chrome__read_page,mcp__claude-in-chrome__get_page_text,mc