playwright-public-web

Inspect rendered UI of approved public websites through an isolated Playwright browser session.

Updated Aug 5, 2026
One-click install
npx skills add https://github.com/ahsanghalib/ai-workflow --skill playwright-public-web-ahsanghalib
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: playwright-public-web
Source: https://github.com/ahsanghalib/ai-workflow/tree/main/skills/playwright-public-web
Command: npx skills add https://github.com/ahsanghalib/ai-workflow --skill playwright-public-web-ahsanghalib

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When you need to gather evidence about how a public website actually renders and behaves, ad-hoc browsing is risky and unstructured. This Skill provides a controlled, read-only workflow for inspecting approved public websites through an isolated browser, with clear approval gates and hard boundaries against state-changing actions. ## Core Features & Use Cases - Read-only UI inspection: Navigate approved origins and capture snapshots, screenshots, and observed behavior without logging in or modifying anything. - Safety boundaries: Enforces strict rules against authentication, form submission, purchases, file uploads, and access to cookies or browser profiles. - Harness-agnostic operation: Works with any isolated browser or Playwright adapter configured by the active AI harness, without assuming a specific CLI or provider. - Use Case: You want to verify how a competitor's public landing page renders its navigation and filters. You approve the exact origin, the Skill inspects the rendered UI, saves screenshots to the project-local artifact directory, and reports observed behavior separately from inference. ## Quick Start Inspect the public website at the approved origin using the isolated browser and report what the rendered UI shows without logging in or changing anything.

Frequently Asked Questions about playwright-public-web

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I inspect a public website with Playwright without logging in?▼

Confirm the exact approved origins and read-only goal, then use the harness's isolated browser profile to navigate and capture snapshots and screenshots. The workflow never authenticates, submits forms, or accesses cookies or storage state.

What browser setup is required before using Playwright for public web inspection?▼

The active harness must have a configured isolated public browser profile created for the project, and the harness must be reloaded or restarted as required. Setup is handled through the harness's documented browser workflow, not inside this Skill.

Can I use this for testing my localhost application?▼

No. This workflow is strictly for unauthenticated public websites. Localhost application testing, login-required sites, and durable regression test suites are explicitly out of scope.

Where are Playwright screenshots saved during inspection?▼

Screenshots are saved only under the configured ignored project-local artifact directory reported by the active harness. The directory name is not fixed; .playwright-mcp/public/ is used only when that is the selected project convention.

What actions are prohibited during public website inspection?▼

Prohibited actions include logging in, creating accounts, submitting forms, making purchases, uploading or downloading files, executing arbitrary page code, and reading browser profiles, cookies, or authentication material. Navigation is limited to approved exact origins.