playwright-trace
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references and is derived from official Playwright resources maintained by a trusted organization. The instructions prioritize using existing project-local dependencies to avoid unnecessary network fetches.- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands via the Playwright CLI to inspect trace files. It specifically uses the --no-install flag to ensure that only pre-installed local packages are used.- [DYNAMIC_EXECUTION]: The skill supports evaluating JavaScript within a browser snapshot context using Playwright's evaluation features. This execution is limited to the browser environment of the trace viewer and is intended for inspecting the DOM state of tested applications.- [INDIRECT_PROMPT_INJECTION]: The skill acts as a processor for external .zip trace files, which could contain untrusted data such as browser logs or page content.
- Ingestion points: Trace data is read into the agent's context through commands like trace open, trace console, and trace snapshot.
- Boundary markers: None explicitly defined to separate trace data from agent instructions.
- Capability inventory: Includes shell command execution and browser-level DOM evaluation.
- Sanitization: Relies on the internal parsing logic of the Playwright CLI tool.
Audit Metadata