centrale-inzichten-lll
Pass
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill follows established patterns for prompt engineering and instructional design. It utilizes local reference files for context and provides structured output to the user. All external links point to legitimate educational resources.
- [PROMPT_INJECTION]: The skill processes user-supplied empathy map input, creating a surface for indirect prompt injection. This is assessed as safe as the instructions focus on data transformation rather than code execution. Ingestion points: User-provided empathy map data input as defined in SKILL.md. Boundary markers: None present. Capability inventory: Writes output files to /mnt/user-data/outputs/ and references the hoe-kunnen-we skill. Sanitization: None present.
Audit Metadata