skills/klotzkette/claude-fuer-deutsches-recht/bdsg-schriftsatz-brief-und-memo-bausteine/Gen Agent Trust Hub
bdsg-schriftsatz-brief-und-memo-bausteine
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 external material provided by the user (such as contracts, administrative acts, and messages) to generate legal drafts. This creates a standard ingestion surface where instructions hidden within the processed documents could potentially influence agent behavior.
- Ingestion points: Reads "Material", "Verwaltungsakte", "Vertragsurkunden", and "Nachrichten" provided by the user as described in SKILL.md.
- Boundary markers: The instructions do not define specific delimiters or "ignore" commands to separate user data from system instructions.
- Capability inventory: The skill's operations are limited to text generation and drafting; no dangerous capabilities such as file system modifications or arbitrary code execution were detected.
- Sanitization: The instructions do not include specific patterns for sanitizing or filtering ingested text content.
- [EXTERNAL_DOWNLOADS]: The skill references and directs the agent to verify information against well-known legal databases.
- Evidence: Instructions include live verification of legal norms via official and well-known services such as gesetze-im-internet.de, dejure.org, and openJur.
Audit Metadata