playwright-cli
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates browsing and scraping third-party web content, which introduces a risk where malicious text on a webpage could influence the agent's behavior. Ingestion points: Page content retrieved via snapshot, eval, or run-code. Boundary markers: The documentation includes warnings advising users to treat extracted text as untrusted. Capability inventory: The skill possesses extensive capabilities including file system writes, network navigation, and arbitrary script execution. Sanitization: Instructions recommend sanitizing data before feeding it back into instructions.
- [DYNAMIC_EXECUTION]: The skill provides the run-code command, which allows for the execution of arbitrary JavaScript using the Playwright API. This is a core functional feature for complex automation but represents a surface for dynamic code execution.
- [COMMAND_EXECUTION]: The skill operates by executing a local command-line tool (playwright-cli) to manage browser sessions and interactions.
Audit Metadata