skills/klotzkette/claude-fuer-deutsches-recht/dsv-pressemitteilung-krisenkommunikation/Gen Agent Trust Hub
dsv-pressemitteilung-krisenkommunikation
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external inputs, including media reports and case documentation, to generate communication products.
- Ingestion points: Case documents (Verwaltungsakte, Protokolle) and active media queries mentioned in SKILL.md.
- Boundary markers: There are no instructions defining boundaries between untrusted input and agent instructions.
- Capability inventory: The skill lacks capabilities for file writing, network transmission, or code execution, significantly limiting the impact of potential injections.
- Sanitization: No input sanitization steps are defined for external data.
- [EXTERNAL_DOWNLOADS]: The instructions recommend verifying legal citations using established legal information portals.
- Evidence: The 'Arbeitsweg' section specifies using external sites (gesetze-im-internet.de, dejure.org, openJur) for live verification of GDPR and BDSG norms.
Audit Metadata