open-source-in-saas
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists exclusively of markdown instructions and metadata. No scripts, binaries, or configuration files are present in the provided skill set.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external legal documents, contracts, and evidence provided by users ("Dokumente und Beweismittel sammeln"). This constitutes an ingestion surface where untrusted data enters the agent context. There are no explicit boundary markers or sanitization steps defined in the instructions. However, since the skill does not define any executable tools, file-system access, or network write capabilities, the risk associated with this surface is minimal.
- [SAFE]: The skill references several external websites for legal research, including gesetze-im-internet.de, dejure.org, and openJur. These are well-known, official, or reputable legal databases used for verifying German and European law and do not represent a security risk.
Audit Metadata