skills/klotzkette/claude-fuer-deutsches-recht/subcontractors-cloud-support-maintenance/Gen Agent Trust Hub
subcontractors-cloud-support-maintenance
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 analyze external documents, creating a surface for indirect prompt injection.
- Ingestion points: The skill directs the agent to analyze administrative acts, contracts, and expert reports within the 'Arbeitsweg' and 'Dokumente und Beweismittel' sections.
- Boundary markers: The instructions lack delimiters or explicit 'ignore' warnings for the processing of these external documents.
- Capability inventory: The skill performs legal research using reputable external sources such as gesetze-im-internet.de and dejure.org. No high-risk capabilities such as command execution or unauthorized file writing are present.
- Sanitization: There are no defined procedures for sanitizing or validating the content of ingested external documents.
Audit Metadata