playwright-cli

Warn

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: MEDIUMDYNAMIC_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill provides commands like run-code and eval that allow the execution of arbitrary JavaScript and Playwright scripts. This functionality enables complex automation but also permits the execution of arbitrary logic within the agent's environment.\n- [COMMAND_EXECUTION]: The skill uses shell commands via npx and npm for installing dependencies and running Playwright tests.\n- [INDIRECT_PROMPT_INJECTION]: The skill's browser automation capabilities create an attack surface for indirect prompt injection from untrusted web pages.\n
  • Ingestion points: Content from web pages retrieved through commands like snapshot, eval, and content.\n
  • Boundary markers: Not used; there are no specific markers to distinguish between instructions and data when processing page content.\n
  • Capability inventory: The skill can write files (state, screenshots, videos), execute arbitrary code, and run shell commands.\n
  • Sanitization: Not performed; the skill does not include steps to sanitize or validate data obtained from the browser.\n- [EXTERNAL_DOWNLOADS]: The skill references the installation of the @playwright/cli package from the NPM registry as a fallback when the global command is missing.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 30, 2026, 02:57 PM
Security Audit — agent-trust-hub — playwright-cli