mandat-triage-it-recht
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted mandate materials (e.g., contracts, documents, facts) which could contain hidden adversarial instructions intended to influence the agent's behavior. \n
- Ingestion points: Mandate documentation and user-provided case facts mentioned in the 'Material' and 'Unterlagen' sections. \n
- Boundary markers: The skill does not define specific delimiters or instructions to ignore instructions embedded within the user-provided materials. \n
- Capability inventory: The skill generates structured triage protocols, aktenanlage, and routes information to secondary legal plugins. \n
- Sanitization: There are no mechanisms described to sanitize or validate the content of the external mandate materials. \n- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to verify legal information using external official websites. \n
- Resources: References well-known, official German and European legal resources including 'gesetze-im-internet.de', 'dejure.org', 'openJur', and official court databases (BVerfG, BGH, EuGH). These references are documented neutrally as they are well-known official services.
Audit Metadata