preliminary-injunction-software
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: Analysis of the skill instructions and metadata revealed no evidence of prompt injection, data exfiltration, obfuscation, or malicious persistence mechanisms. The skill primarily directs the agent to perform legal research and document analysis.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external legal documents, which constitutes an ingestion surface for potential indirect instructions hidden in those files.
- Ingestion points: The skill ingests 'Verwaltungsakte, Vertragsurkunden, Schriftsätze, Bescheide, Protokolle, Sachverständigengutachten' (SKILL.md).
- Boundary markers: There are no explicit boundary markers or directives to ignore instructions embedded in the analyzed documents.
- Capability inventory: The skill's capabilities are limited to generating legal outputs like memos, matrices, and briefings. It does not perform network operations or sensitive file system modifications.
- Sanitization: No specific sanitization or escaping mechanisms for external content are defined in the instructions.
Audit Metadata