skills/matteocervelli/llms/frontend/Gen Agent Trust Hub

frontend

Pass

Audited by Gen Agent Trust Hub on Aug 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The DesignSystemFetcher tool in tools/design_system_fetcher/fetcher.py downloads content from remote URLs to extract design tokens such as colors and typography.
  • [DATA_EXFILTRATION]: The /frontend review command in SKILL.md is designed to send screenshots and source code to the Google Gemini API for visual design analysis.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes design documentation from external URLs; while intended for structured token extraction, this ingestion represents a potential surface for instructions embedded in source data.
  • [COMMAND_EXECUTION]: The skill uses design-lint.sh to perform static analysis of project files using grep patterns to detect design anti-patterns.
  • [DYNAMIC_EXECUTION]: The fetcher.py script dynamically modifies the Python search path using sys.path.insert to resolve local module dependencies.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 29, 2026, 08:31 AM
Security Audit — agent-trust-hub — frontend