Publisher: OpenAI · Editorial profile by AI Tools Ratings · Review method
Not runtime tested; source-document review only. No security certification.
What task is this skill designed for?
Reproduce a failed search flow on a staging website and capture the visible result after entering a query and submitting the form.
What input and output should you expect?
Input: A staging URL, a short reproduction sequence, a test account if needed and a clear limit on which actions may be submitted.
Output: Expected, not independently tested: a browser interaction trace or screenshots and a description of the reproduced state. The source is a CLI workflow, not automatically a committed regression-test suite.
What does it need to run?
Node.js/npm with npx, a supported browser runtime and permission to use the bundled Playwright CLI wrapper. Package retrieval may need internet access. Test credentials must be provisioned through the local environment rather than embedded in a prompt.
Authored for Codex in the OpenAI skills repository. This directory has not independently installed or executed the skill.
How should you approach installation?
- Read the pinned CLI workflow and wrapper before installing dependencies.
- Confirm Node/npm and the browser are available; use current package and agent documentation rather than copying an old global installation path.
- Run a harmless staging navigation and inspect its snapshot before allowing form submission.
Installation guide: discovery, dependencies and runtime checks
What files, accounts or data can it access?
Can navigate pages, read page content, fill forms and trigger website actions using the active browser session. A logged-in session may allow writes or purchases; restrict first trials to staging and reversible test operations.
What does the skill cost?
The skill source does not list a purchase price. Agent usage, hosted browsers and third-party websites may carry separate costs.
What are its limits and alternatives?
Element references can become stale after navigation or DOM changes. A successful click does not prove the server accepted a request. Native desktop dialogs and some browser permission prompts require a different interaction surface.
Use a maintained Playwright test suite when the goal is repeatable CI coverage. Use this CLI skill for interactive investigation or a concrete browser task with observable results.
What could you ask in a small trial?
Try this with a small sample:
On the staging search page, reproduce the no-results state using the query test-example. Inspect the current page before clicking, record the result, and do not create accounts or change data.
How would you verify the result?
- Confirm the final URL and visible result match the requested search.
- Check screenshots and captured logs for account details before sharing.
- If a product fix is made, add an appropriate regression check separately instead of claiming the exploratory run is CI coverage.
Use a baseline and record acceptance evidence before treating a successful task as a reliable workflow.
Which source and license were checked?
Read the reviewed SKILL.md at its pinned commit · Complete reviewed folder
Repository and current publisher guidance
Checked:
Apache-2.0 — identified in skills/.curated/playwright/LICENSE.txt at the reviewed commit. Third-party assets and dependencies may have separate terms. License evidence
Version details and review scope
Source: openai/skills/skills/.curated/playwright
Reviewed commit: 49f948faa9258a0c61caceaf225e179651397431
SKILL.md SHA-256: 0ffaabcc8e0990627c4725f18bf1c7955534a796c1c199e872909de2013ce6a8
Source-document review. Source text and stated dependencies were inspected for this profile. Referenced code was not comprehensively audited, no runtime test was performed, and source files are not mirrored here.