amtsangemessene-alimentation
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection through the ingestion of external legal documents and administrative records.
- Ingestion points: The skill is instructed to collect and analyze documents such as 'Ernennungsurkunde', 'dienstliche Beurteilung', 'Disziplinarverfügung', and 'Versorgungsbescheid' as listed in SKILL.md.
- Boundary markers: There are no explicit delimiters or specific instructions provided to the agent to treat content within these documents as untrusted data or to ignore embedded instructions.
- Capability inventory: The agent uses the analyzed data to generate 'Handlungsempfehlungen' (action recommendations), legal drafts, and 'Risikomatrizen' (risk matrices) as described in SKILL.md.
- Sanitization: No sanitization or validation mechanisms are defined for the content of the ingested files.
- [SAFE]: The skill references well-known and official legal sources such as gesetze-im-internet.de, dejure.org, and openJur. No automated network execution or unauthorized sensitive file access was identified.
Audit Metadata