ontwerper-digitaal-product
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze user-provided code, descriptions, and screenshots, creating a vector for indirect prompt injection.\n
- Ingestion points: The prompt in SKILL.md explicitly directs the agent to process user-shared code, screenshots, and descriptions.\n
- Boundary markers: The instructions lack explicit delimiters or safety warnings to ignore instructions embedded within the processed user data.\n
- Capability inventory: The skill is configured with
Read,Write,Edit, andBashtools, providing a surface for file system and command operations if an injection occurs.\n - Sanitization: There is no evidence of input validation or sanitization for external content.\n- [COMMAND_EXECUTION]: The skill uses a local script
scripts/contrast_checker.pyto perform WCAG contrast checks.\n - Evidence: The script is a standard utility that uses only the
syslibrary to process hex color codes and calculate relative luminance. It is executed via theBashtool as documented in the instructions.
Audit Metadata