redesign-existing-projects

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional and provides design guidelines for refactoring web applications. No executable scripts, binaries, or suspicious commands were found.- [EXTERNAL_DOWNLOADS]: The skill references the well-known placeholder image service picsum.photos. This is a standard resource for UI development and is considered safe.- [DATA_EXFILTRATION]: No indicators of sensitive data harvesting or unauthorized exfiltration to external domains were detected.- [PROMPT_INJECTION]: The skill does not contain instructions to override system safety filters or engage in role-play scenarios designed to bypass constraints.- [COMMAND_EXECUTION]: There are no instructions to run privileged or dangerous shell commands. All suggested actions focus on CSS and code-level refactoring.- [SAFE]: Regarding indirect prompt injection: Ingestion point: The skill scans existing codebase files (SKILL.md). Boundary markers: None. Capability inventory: Reads and modifies codebase files (SKILL.md). Sanitization: Instructions emphasize maintaining existing functionality and reviewability. This surface represents a minimal risk as the skill is focused on design improvements.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 01:13 AM
Security Audit — agent-trust-hub — redesign-existing-projects