plugins/src/expo/skills/ops-browser-uat/SKILL.md
Browser-based user acceptance testing via Playwright MCP tools. Logs into the application, navigates through features, and captures visual proof with screenshots.
npx skillsauth add codyswanngt/lisa ops-browser-uatInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Perform browser-based user acceptance testing using Playwright MCP tools.
Argument: $ARGUMENTS — scenario and environment (e.g., smoke-test dev, login staging, player-detail dev, custom production /path)
playwright browser and load all browser tools.Before running any scenario, read these project files to discover configuration:
e2e/constants.ts — environment URLs, test credentials (phone, OTP), timeouts, viewportse2e/selectors.ts — all data-testid values organized by feature areae2e/fixtures/auth.fixture.ts — exact login/logout flow to replicate via MCP toolsplaywright.config.ts — video/trace recording config, test directoryReplicate the login flow from e2e/fixtures/auth.fixture.ts using Playwright MCP tools. Read the file to get the exact steps, which typically follow this pattern:
browser_navigate to {BASE_URL}/signinbrowser_snapshot to discover page elementsbrowser_fill_form — fill the phone input (discover placeholder text from auth fixture)browser_click — click the "Next" buttonbrowser_wait_for — wait for URL to contain confirm-code (timeout from constants)browser_snapshot to discover OTP inputbrowser_fill_form — fill the OTP input (discover testid from selectors, value from constants)browser_wait_for — wait for URL to NOT contain confirm-codebrowser_take_screenshot — capture login proofExpo's development error overlay may appear. Dismiss it:
browser_evaluate with script:
document.getElementById('error-overlay')?.remove();
document.querySelectorAll('[id*="error"]').forEach(el => el.remove());
browser_press_key — press "Escape" as backup// browser_evaluate
localStorage.getItem("@authData") !== null
Discover the menu button and sign-out flow from e2e/selectors.ts and e2e/fixtures/auth.fixture.ts:
browser_click — menu button element (discover testid from selectors)browser_click — element with text "Sign Out"browser_wait_for — URL contains signinRead e2e/selectors.ts to get the complete data-testid reference for the project. The selectors object is organized by feature area (auth, nav, home sections, player detail, etc.).
Use [data-testid="{TESTID}"] to target elements in Playwright MCP tools.
Full app walkthrough. Read e2e/selectors.ts to identify which sections and elements to verify:
Just the login flow with screenshot proof.
Login → verify all home screen sections have data. Screenshot each section.
Login → navigate to user-specified URL → interact as instructed → screenshot.
At each verification point:
browser_take_screenshot — visual proof of current statebrowser_console_messages — check for JavaScript errorsbrowser_network_requests — monitor for failed API calls (4xx, 5xx)Report UAT results as a table:
| Step | Action | Status | Notes | |------|--------|--------|-------| | 1 | Navigate to /signin | PASS | Page loaded in 1.2s | | 2 | Enter phone number | PASS | Input accepted | | 3 | Click Next | PASS | Navigated to /confirm-code | | 4 | Enter OTP | PASS | OTP accepted | | 5 | Verify home screen | PASS | All sections present | | 6 | Logout | PASS | Redirected to /signin |
Include total PASS/FAIL count and any JS errors or failed network requests observed.
development
Prepare a machine — a fresh laptop or a throwaway container — to run coding agents, before any repository exists. Detects which of Lisa's supported agents (Claude Code, Codex, Cursor, OpenCode, Antigravity, Copilot) are already installed, asks which credential manager the machine uses (Bitwarden, 1Password, Doppler, Vault, AWS, or none), and installs only what is missing, each by its vendor's own preferred method. Idempotent, headless by default, and emits a Dockerfile for a spin-up/spin-down environment. Run it on a new machine, in a container, or before cloning anything.
tools
Provision and verify a remote execution environment for a host project — Codex Cloud today, other remote surfaces as they are added. Generates a repository-owned setup script that installs the declared toolchain, materializes secrets through lisa-secrets-access, and runs the project's own hook. Provisions by API where one exists, by driving the vendor console where one does not, and by emitting exact config otherwise — then proves the result with the same read-back regardless of which tier did the work. Use before dispatching any work with executionEnv.
tools
Bring a developer's machine in line with the toolchain the project declares. Reports every tool in remoteEnv.tools that is missing, outdated, or unpinned for this platform, and installs the missing ones into ~/.local/bin from the same pinned, checksummed entries the remote surfaces use — but only when asked. Same manifest, same pins, same installers as lisa-setup-remote-env; what differs is consent and that the pin is a floor rather than an equality. Run it on a fresh checkout, after a manifest change, or when a tool fails at the moment of use.
tools
Route one unit of work to a remote execution surface. Reads the executionEnv parameter (local by default, codex-cloud or claude-web today), verifies the environment is provisioned and bound to this repository, submits a thin skill invocation, records the task identifier to .lisa/remote-dispatch.json, and exits without polling. Routing only — the remote runs the identical skill from the identical repository. Composable and inline: other skills invoke it via the Skill tool rather than users calling it directly.