authority-check
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided legal documents and citations, creating a surface for indirect prompt injection if the processed content contains adversarial instructions.
- Ingestion points: The skill accepts untrusted user input via the
argument-hintparameter and processes text for extraction in step 1. - Boundary markers: The instructions lack explicit boundary markers or delimiters to isolate user-supplied data from the agent's instructions.
- Capability inventory: The skill is instructed to call external legal databases and utilize Model Context Protocol (MCP) tools for verification.
- Sanitization: No input sanitization, filtering, or validation mechanisms are mentioned in the skill instructions.
Audit Metadata