playwright
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing Playwright and associated tools like axe-core using standard package managers (npm, pnpm, yarn). It also references official Docker images hosted on mcr.microsoft.com. These are verified sources for the documented framework.
- [COMMAND_EXECUTION]: The documentation includes various CLI commands for running tests, managing browser binaries, and starting remote servers (e.g., npx playwright test, docker run). These are essential operational commands for the framework's intended use.
- [INDIRECT_PROMPT_INJECTION]: The section on 'Playwright Test Agents' describes AI-powered components (Planner, Generator, Healer) that ingest user requests and requirements to produce code. This functionality presents a potential attack surface if untrusted data is processed, although the documentation describes standard developer workflows.
- [DYNAMIC_EXECUTION]: The skill documents the use of page.evaluate() and page.addInitScript(), which allow the execution of arbitrary JavaScript within the browser context. This is a core feature of the automation framework used for page interaction and state management.
Audit Metadata