dsv-schnelltriage-risiko
Pass
Audited by Gen Agent Trust Hub on Aug 23, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to process untrusted external data such as contracts, administrative acts, and expert reports to perform risk assessments, which constitutes an indirect prompt injection surface.\n
- Ingestion points: Processes materials including 'Verwaltungsakte', 'Vertragsurkunden', and 'Sachverständigengutachten' as listed in SKILL.md.\n
- Boundary markers: No delimiters or specific instructions to ignore embedded commands within the processed content are present.\n
- Capability inventory: The skill is limited to generating text-based triage memos and does not have access to tools for file system modification, command execution, or network exfiltration.\n
- Sanitization: There are no specified procedures for validating or filtering the content of the external documents.\n- [EXTERNAL_DOWNLOADS]: The skill verifies legal information through established public resources.\n
- Evidence: Accesses 'gesetze-im-internet.de', 'dejure.org', 'openJur', and court databases (BVerfG, BGH, EuGH) to verify legal norms and case law (SKILL.md).
Audit Metadata