skills/klotzkette/claude-fuer-deutsches-recht/persoenlichkeitsrecht-stimme-deepfake/Gen Agent Trust Hub
persoenlichkeitsrecht-stimme-deepfake
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks were detected. The skill consists of text-based instructions for legal research and analysis.
- [EXTERNAL_DOWNLOADS]: The skill directs the agent to verify legal information using official and well-known German and European legal databases, such as gesetze-im-internet.de, dejure.org, and eur-lex.europa.eu. These references to authoritative legal sources are expected for the skill's primary purpose and do not represent a security risk.
- [INDIRECT_PROMPT_INJECTION]: The skill includes instructions to ingest and process user-provided documents and evidence (e.g., contracts, files, and metadata). While this creates an attack surface for indirect prompt injection, the skill does not grant the agent any dangerous capabilities such as file writing, shell execution, or unauthorized network operations, making the surface non-exploitable within the context of these instructions.
Audit Metadata