keine-unnoetigen-aussagen
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to collect and analyze external evidence such as contracts (Werkvertrag/Dienstvertrag), tax documents (Steuererklärung), and applications (Statusfeststellungsantrag). This creates an attack surface where malicious instructions could be embedded in those documents to influence the agent's behavior.
- Ingestion points: The 'Dokumente und Beweismittel sammeln' section in SKILL.md.
- Boundary markers: None; the instructions do not specify how to delimit external content.
- Capability inventory: No active code or tool-specific definitions are included in the skill; it relies on the agent's default capabilities.
- Sanitization: Not present.
Audit Metadata