| name | pp-validate |
| description | Slash command — run the Power Platform Playwright sample suite. Walks the user through prerequisites, .env, auth state, and a chosen Playwright project, then parses results and explains failures using CLAUDE.md anti-patterns. Wraps the pp-playwright-validate agent. |
/pp-validate — run the Power Platform Playwright sample suite
When the user types /pp-validate, invoke the pp-playwright-validate agent and pass any arguments through.
Argument forms accepted
| Form | Action |
|---|
/pp-validate | Full interactive run — agent collects .env answers, runs all projects |
/pp-validate canvas-app | Run only the canvas-app Playwright project |
/pp-validate canvas-app --headed | Same, with visible browser |
/pp-validate --skip-auth | Reuse existing .playwright-ms-auth/state-*.json files (24h window) |
/pp-validate model-driven-app --headed --skip-auth | All flags compose |
Valid project names: canvas-app, model-driven-app, custom-page, studio-authoring, gen-ux-runtime, default, or all (default).
Behaviour
- Spawn the pp-playwright-validate agent via the Agent tool.
- Pass the parsed argument set as the agent's initial brief.
- The agent owns Steps 0–7 (prereqs →
.env → auth → run → parse → explain → report).
- Surface the agent's final result banner + failure list back to the user.
When NOT to use
- For test authoring use
/pp-author.
- For analysing an existing failure (no re-run) use
/pp-diagnose.
