Legal Client Intake
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and summarize untrusted data provided by potential clients during the intake process.\n
- Ingestion points: Prospect descriptions of legal matters, party names, and case facts collected during interactive dialogue as specified in SKILL.md.\n
- Boundary markers: The skill provides templates for structured intake (e.g., 'INTAKE SUMMARY'), but lacks explicit instructions to treat prospect input as untrusted data or to ignore instructions that might be embedded within a prospect's narrative.\n
- Capability inventory: The skill documentation mentions potential integration with legal practice management software such as Clio, MyCase, and PracticePanther to create matter records directly from intake data.\n
- Sanitization: There is no mention of sanitization, escaping, or validation of the input provided by prospects before it is used to generate summaries or populate external records.
Audit Metadata