playwright-cli
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides the
run-codecommand, which allows the execution of arbitrary JavaScript code within the browser context. This functionality is core to the tool's purpose, facilitating tasks like permission management, geolocation emulation, and complex DOM interactions.\n- [DATA_EXFILTRATION]: The skill includes capabilities to retrieve and export sensitive browser information, including cookies (cookie-get,cookie-list), localStorage contents, and full session state files (state-save auth.json). While standard for browser automation, these tools handle sensitive user credentials.\n- [EXTERNAL_DOWNLOADS]: The documentation recommends the installation of the@playwright/clipackage from the NPM registry. This is the official package for the Playwright framework and represents a standard, trusted dependency.\n- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes content from untrusted external websites through snapshots, console output, and element evaluations.\n - Ingestion points: Page snapshots, element evaluation via
eval, and browser console logs.\n - Boundary markers: None. The skill does not provide instructions to the agent to delimit or ignore instructions found within web page content.\n
- Capability inventory: The skill can execute shell commands, write files (screenshots, videos, auth states), and run arbitrary JavaScript.\n
- Sanitization: No sanitization of ingested web content is described.
Audit Metadata