ui-cloner-quality-check

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill implements an indirect prompt injection surface by processing and interpolating data from external project files into its final output instructions.\n
  • Ingestion points: The skill reads site metadata and design philosophy from 'plans/01-site-dna.md' to verify and patch the generated prompt.\n
  • Boundary markers: The instructions do not define delimiters or protective prompts to prevent instructions embedded within 'plans/01-site-dna.md' from influencing the agent's behavior during the 'Failure Protocol' phase.\n
  • Capability inventory: The skill is capable of reading local file content and writing results to 'plans/04-final-prompt.md'.\n
  • Sanitization: No validation or escaping of the ingested site data is performed before it is used to generate the final design system specifications or the 'Execution directive'.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 05:19 AM