playwright-best-practices
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill naturally interacts with external data sources such as Playwright trace files, CI reports, and application source code. It includes robust defensive instructions requiring the agent to treat all artifact content as evidence rather than commands, specifically designed to mitigate indirect prompt injection.
- [DATA_EXFILTRATION]: The skill includes clear guidance on data privacy, instructing the agent to redact credentials and sensitive user data from any excerpts or reports generated during troubleshooting.
- [DYNAMIC_EXECUTION]: The troubleshooting workflow utilizes the Playwright CLI's native capability to evaluate JavaScript within page snapshots. This is documented as a standard debugging technique and is restricted to the context of the reconstructed historical page.
- [EXTERNAL_DOWNLOADS]: All external libraries mentioned in the documentation (e.g., axe-core, web-vitals) are well-known, industry-standard packages from established registries or CDNs like unpkg.com, and are used solely for legitimate testing purposes.
Audit Metadata