verbotene-ki-praktiken
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes untrusted descriptions of AI systems provided by users to determine legal compliance. It lacks explicit boundary markers or instructions to the agent to disregard instructions potentially embedded within the input data, creating a potential surface for indirect prompt injection.
- Ingestion points: Input fields in SKILL.md for system function, purpose, and context.
- Boundary markers: Not present in the instructions.
- Capability inventory: The skill itself does not define any executable code or tool-access capabilities.
- Sanitization: No sanitization or filtering of input content is specified.
- [NO_CODE]: The skill consists entirely of instructional markdown and descriptive metadata. It does not include any scripts, binaries, or executable configuration files for analysis.
Audit Metadata