skills/klotzkette/claude-fuer-deutsches-recht/agg-pruefung-bewerber-und-beschaeftigte/Gen Agent Trust Hub
agg-pruefung-bewerber-und-beschaeftigte
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data such as job advertisements, correspondence, and case narratives provided by the user, which could contain malicious instructions.
- Ingestion points: User-provided inputs defined in the 'Eingaben' section, specifically
Sachverhaltschilderung(case description) andBeweismittel(evidence). - Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the processed data.
- Capability inventory: The skill produces structured text outputs, such as legal matrices and draft correspondence, but does not perform shell execution or external network writes based on the data.
- Sanitization: No sanitization or validation logic for the external user-provided content is specified.
- [SAFE]: The skill fetches legal information from established and trusted official domains including gesetze-im-internet.de, dejure.org, openjur.de, and bundesarbeitsgericht.de.
- [SAFE]: The skill references a local configuration file located at
~/.claude/plugins/config/claude-fuer-deutsches-recht/arbeitsrecht/CLAUDE.md, which is part of the plugin's own environment for retrieving HR guidelines.
Audit Metadata