kanzleitag-simulation-kanzlei
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data from multiple channels including E-mail, Messenger, Fax, and ZIP archives, which could contain malicious instructions.
- Ingestion points: External data enters the agent's context during the intake and mail processing phases (10:00 and 11:00) as defined in the Eight-Hour-Schedule in
SKILL.md. - Boundary markers: There are no specific delimiters or warnings instructing the agent to ignore or isolate instructions found within the processed external data.
- Capability inventory: The skill claims to handle high-privilege operations such as legal electronic communication (beA), payroll (SV-Meldung), and tax transmissions (UStVA).
- Sanitization: The instructions do not define any validation, escaping, or filtering mechanisms for data ingested from external sources.
Audit Metadata