anfaenger-workflow-sozialgericht
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted user materials, such as legal notices (Bescheid), expert opinions (Gutachten), and court rulings. This creates a surface for indirect prompt injection where malicious instructions hidden within these documents could attempt to influence the agent's analysis or output.
- Ingestion points: Material ingested for triage and analysis as described in the 'Erste Triage' section of SKILL.md.
- Boundary markers: The skill lacks explicit boundary markers or instructions for the agent to ignore embedded commands within the processed materials.
- Capability inventory: The skill's primary capabilities are text-based generation of legal summaries, risk assessments, and procedural drafts. It does not appear to utilize high-privilege tools for network access or filesystem modification.
- Sanitization: There is no mention of input sanitization, filtering, or validation for the content of the documents being processed.
Audit Metadata