playwright
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- Prompt Injection (SAFE): No instructions attempting to override agent behavior, bypass safety filters, or extract system prompts were detected.
- Data Exposure & Exfiltration (SAFE): No sensitive file paths (e.g., SSH keys, environment variables) or hardcoded credentials were found. The skill correctly recommends adding the 'playwright/.auth/' directory to .gitignore to prevent accidental exposure of session data.
- Unverifiable Dependencies & Remote Code Execution (SAFE): All referenced packages (e.g., @playwright/test, msw, next) are from trusted sources. Commands like 'npm run build' and 'npx playwright test' are standard for the development and testing lifecycle.
- Indirect Prompt Injection (LOW): As a reference guide, the skill discusses interacting with web content, but it does not provide a script that ingests untrusted data. The capability to execute browser-side code via 'page.evaluate' is a standard testing feature and not used maliciously here.
- Obfuscation (SAFE): All content is provided in clear, human-readable markdown and typescript. No Base64, zero-width characters, or homoglyphs were found.
Audit Metadata