redesign-existing-projects-v2

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill consists of instructional content and design guidelines aimed at UI/UX improvement. No evidence of obfuscation, malicious command execution, or unauthorized file access was found.\n- [EXTERNAL_DOWNLOADS]: The skill references the well-known service picsum.photos to fetch placeholder images during the redesign process.\n- [EXTERNAL_DOWNLOADS]: The skill metadata and provenance documents point to a community repository on GitHub (sickn33/antigravity-awesome-skills) as the original source of the workflow instructions.\n- [PROMPT_INJECTION]: The skill is designed to scan and audit existing codebases, which creates an indirect prompt injection surface. This is considered low risk as the skill's capabilities are restricted to visual styling and UI component adjustments.\n
  • Ingestion points: The skill reads existing frontend code during the 'Scan' phase described in SKILL.md.\n
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the guidelines.\n
  • Capability inventory: The agent is authorized to modify CSS, HTML structure, and UI components. It does not possess network exfiltration or shell execution capabilities.\n
  • Sanitization: No sanitization or validation of the processed codebase is specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 11:21 PM
Security Audit — agent-trust-hub — redesign-existing-projects-v2