skills/klotzkette/claude-fuer-deutsches-recht/software-scope-und-rechtsgebiet-router/Gen Agent Trust Hub
software-scope-und-rechtsgebiet-router
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external legal documents which are potential vectors for indirect prompt injection. * Ingestion points: The skill collects external evidence including 'Vertragsurkunden' (contracts), 'Schriftsätze' (legal briefs), and 'Bescheide' (notices) as described in the 'Arbeitsweg' section of SKILL.md. * Boundary markers: There are no instructions defining delimiters to separate external text from the agent's internal instructions. * Capability inventory: The skill uses analysis and report generation capabilities; no direct file system or shell execution is specified. * Sanitization: No filtering or sanitization of input from external sources is described.
Audit Metadata