Scaffold e2e

/scaffold-e2e

Single-backend skill. Calls scaffold_e2e on the rolepod-uiproof MCP server. No fallback (D-024).

When to use

• The user asks to generate an e2e test for a flow they just verified interactively.
• A replay bundle from /verify-ui exists and should be transcribed into a real test file.
• A mobile target (iOS / Android / React Native / Flutter) needs a test scaffold — use framework: "maestro" to emit a Maestro YAML flow.

When NOT to use

• A unit or integration test is sufficient — pick a closer framework manually.
• The scenario is too vague to scaffold — ask the user to clarify before calling.

Coverage

The codegen handles every step kind and expect kind supported by /verify-ui (click, type, key, navigate, wait_for, hover, drag, fill_form, upload, dialog, set_env, switch_page, evaluate; text_visible, text_absent, url_matches, ref_in_state, no_console_errors, no_failed_requests, request_made, response_status).

Playwright-test gets first-class translation for everything that has a direct Playwright API. Pytest+selenium covers the basics; expect kinds that need network introspection (no_failed_requests, request_made, response_status) emit a TODO referencing selenium-wire or BiDi, since upstream Selenium has no network-capture primitive.

Maestro (mobile) emits a YAML flow (launchApp / tapOn / inputText / assertVisible / extendedWaitUntil / scroll), following Maestro's own conventions rather than a translation of the Playwright template. Web-only kinds (hover, switch_page, evaluate, network expects) degrade to comments instead of failing. TC-ID traceability: a TC<n> id and P1/P2 priority found in scenario_nl are carried into the filename (TC2_<slug>.yaml), the header comment, and Maestro tags — greppable exactly like the web frameworks' test names. The Maestro CLI is NOT required to scaffold (only to run); the scaffold hands off — it never auto-runs.

Inputs

• framework — playwright-test | vitest+playwright | pytest+selenium | maestro.
• scenario_nl — natural-language description of the scenario. For maestro, lead with the TC id and priority (e.g. "TC2 [P1] …") to preserve traceability.
• url — entry URL. Required for the web frameworks; for maestro it marks a web flow instead of app_id.
• app_id (maestro only) — application id of the mobile app under test (e.g. com.example.app).
• recorded_bundle (optional) — path to a replay bundle from a prior /verify-ui run; when present, steps and expectations are transcribed.
• filename (optional) — override the generated file name.

Outputs

• run_id — folder under ./.rolepod-uiproof/artifacts/.
• test_file_path — path to the generated test file.
• language — typescript | python | yaml.
• dependencies — packages the user needs to install (empty for maestro — the CLI is an external binary, pointed to in setup_notes).
• setup_notes — what to run after install (for maestro: maestro test <flow.yaml>, run by the caller, never by the tool).
• from_replay_bundle — boolean indicating whether the file was transcribed from a recorded run.

Process

1. Build scaffold_e2e input.
2. Call the tool.
3. Print the generated file path and the setup steps. Surface dependencies as an install command.

Evidence routing

Run artifacts (the generated test file) are saved under:

• Standalone: .rolepod-uiproof/artifacts/<prefix>_<ts>_<uuid>/
• With rolepod parent (detected via the marker file <git-root>/.rolepod/parent-active written by the parent's SessionStart hook): <git-root>/.rolepod/evidence/<ts>-rolepod-uiproof-<skill>/

Either way the run directory contains a manifest.json per Extension Protocol v1 (with phase: "build" for this skill).

If the tool is unavailable

Surface plainly:

The /scaffold-e2e skill needs the rolepod-uiproof MCP server, which is not currently available. Confirm the plugin is installed and try again.

Do not attempt another backend (D-024).

Examples

Transcribe a replay bundle to a Playwright Test file

{
  "framework": "playwright-test",
  "scenario_nl": "user opens example.com and clicks Learn more",
  "url": "https://example.com",
  "recorded_bundle": ".rolepod-uiproof/artifacts/verify_…/replay.json"
}

Returns:

{
  "run_id": "scaffold_…",
  "test_file_path": ".rolepod-uiproof/artifacts/scaffold_…/user-opens-example-com-and-clicks-learn-more.spec.ts",
  "language": "typescript",
  "dependencies": ["@playwright/test"],
  "setup_notes": "Install: npm i -D @playwright/test && npx playwright install. Run: npx playwright test.",
  "from_replay_bundle": true
}

Scaffold a Maestro flow for a mobile test case

{
  "framework": "maestro",
  "scenario_nl": "TC2 [P1] user logs in with minimum boundary password",
  "app_id": "com.example.app"
}

Returns test_file_path ending in TC2_user-logs-in-with-minimum-boundary-password.yaml with the TC id in the header comment and tags, language: "yaml", empty dependencies, and setup_notes pointing at maestro test <flow.yaml> (hand-off — the caller runs it).