playwright-cli

Pass

Audited by Gen Agent Trust Hub on May 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the playwright-cli tool to automate browser behavior and perform file system operations such as saving screenshots and PDFs.
  • [REMOTE_CODE_EXECUTION]: Commands such as eval and run-code allow the agent to execute arbitrary JavaScript code within the browser context, providing a mechanism for dynamic code execution.
  • [DATA_EXFILTRATION]: The skill has the capability to access and export sensitive session information, including cookies, local storage, and authentication states (e.g., via cookie-list and state-save).
  • [PROMPT_INJECTION]: The skill navigates to and processes content from arbitrary external websites using playwright-cli goto, creating a surface for indirect prompt injection attacks.
  • Ingestion points: External web content retrieved via the goto command.
  • Boundary markers: None; the instructions do not specify any delimiters or warnings to ignore instructions embedded in page content.
  • Capability inventory: Shell command execution, network access, and file writing.
  • Sanitization: No validation or sanitization of page content is described.
Audit Metadata
Risk Level
SAFE
Analyzed
May 22, 2026, 04:18 PM
Security Audit — agent-trust-hub — playwright-cli