testing-data-visualizations
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill inherently possesses an attack surface for indirect prompt injection as it is designed to ingest and process data for testing visualizations. This is a functional requirement of the skill and not a malicious finding.
- Ingestion Points: The skill processes external data via mocked API responses and fixture files, as documented in 'assets/templates/playwright-visual-regression-starter.ts' and 'references/data-mocking-and-fixtures.md'.
- Boundary Markers: The skill explicitly advocates for mocking at the network or repository boundary to maintain a separation between data logic and rendering, as seen in the 'Working Pattern' section of 'SKILL.md'.
- Capability Inventory: The provided templates utilize the Playwright framework for browser automation, network interception, and visual regression (screenshot) capabilities.
- Sanitization: The 'Key Takeaways' in 'references/data-mocking-and-fixtures.md' specifically recommend that any production-derived fixtures be kept 'small, sanitized, and versioned'.
Audit Metadata