visual-regression
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No security threats were detected in the skill instructions or metadata. The skill follows industry-standard practices for visual testing.
- [EXTERNAL_DOWNLOADS]: References the microsoft/playwright framework for browser automation. This is a well-known, trusted repository from a recognized organization and is considered safe practice for web UI testing.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted web pages, which presents an attack surface for indirect prompt injection.
- Ingestion points: Web pages rendered and captured via Playwright in the baseline and re-capture steps.
- Boundary markers: No specific boundary markers are defined for the visual content processed.
- Capability inventory: The skill utilizes browser automation (Playwright) and local server execution via helper scripts.
- Sanitization: The methodology explicitly mandates human review of all differences and requires written confirmation before updating any baselines, which provides strong mitigation against automated instruction following from processed data.
Audit Metadata