web-design-guidelines

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • Remote Resource Integration: The skill fetches guidelines from an external GitHub repository. This allows the tool to use the most recent rules provided by the author to ensure design compliance.\n- Dynamic Instruction Context: The skill retrieves content from a remote URL that informs its auditing logic.\n
  • Ingestion points: Guidelines are retrieved from the GitHub URL specified in SKILL.md.\n
  • Boundary markers: No specific markers are used to delimit external instructions within the skill.\n
  • Capability inventory: The skill utilizes WebFetch for retrieval and file reading capabilities for auditing.\n
  • Sanitization: No explicit sanitization or filtering of the remote content is described in the skill instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 01:10 AM
Security Audit — agent-trust-hub — web-design-guidelines