qe-security-visual-testing

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external websites through browser navigation.
  • Ingestion points: The skill accepts a --url parameter in its primary CLI workflows and via the scanUrl method in the integration snippets within SKILL.md.
  • Boundary markers: The instructions explicitly require the agent to "VALIDATE URLs before navigation" and "SCAN for PII before saving screenshots" to prevent processing of malicious schemes or storage of sensitive data.
  • Capability inventory: The skill utilizes CLI tools and SDKs to capture screenshots, store baselines, and generate reports in the aqe/security-visual/ memory namespace, as documented in the Workflows and Memory Namespace sections of SKILL.md.
  • Sanitization: It implements active sanitization by blocking specific malicious URL schemes (javascript:, data:, file:) and providing automated PII masking strategies such as blurring, redaction, and pixelation.
  • [EXTERNAL_DOWNLOADS]: The skill documentation references external dependencies for its operation.
  • The skill relies on the @agentic-qe/v3 Node.js package for integration and the aqe command-line utility for testing pipelines, neither of which are identified as standard global libraries or recognized trusted vendor resources.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 06:15 PM
Security Audit — agent-trust-hub — qe-security-visual-testing