skills/klotzkette/claude-fuer-deutsches-recht/dekompilierung-interoperabilitaet-devops/Gen Agent Trust Hub
dekompilierung-interoperabilitaet-devops
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 ingest and analyze untrusted external data sources such as administrative acts (Verwaltungsakte), contracts (Vertragsurkunden), pleadings (Schriftsätze), and expert reports (Sachverständigengutachten).
- Ingestion points: The workflow in
SKILL.mdexplicitly directs the agent to collect and analyze various external legal documents and case files. - Boundary markers: The instructions do not specify the use of delimiters (e.g., XML tags or triple backticks) to separate untrusted document content from the agent's instructions.
- Capability inventory: The skill instructs the agent to perform "live checks" on external legal databases (gesetze-im-internet.de, dejure.org, openJur), which implies the use of web search or browsing tools.
- Sanitization: There are no specific instructions to sanitize, escape, or ignore embedded instructions within the ingested documents, which could allow adversarial content in a contract or pleading to influence the agent's reasoning.
Audit Metadata