ai-legal-ops
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFENO_CODEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill contains no executable scripts, binaries, or command-line tools. It functions as a set of natural language instructions guiding the agent's behavior.
- [EXTERNAL_DOWNLOADS]: The instructions direct the agent to verify legal information using official government and established databases including gesetze-im-internet.de, dejure.org, and eur-lex.europa.eu. These represent trusted and necessary sources for legal research.
- [INDIRECT_PROMPT_INJECTION]: The skill describes a data ingestion surface where the agent processes legal documents, evidence, and financial metrics (SKILL.md). While boundary markers and sanitization are not explicitly defined, the skill lacks dangerous tools or network exfiltration capabilities that would allow for the exploitation of this surface. Furthermore, the instructions mandate human review and verification of all findings against primary sources.
Audit Metadata