skills/box/skills/box-legal-workflows/Gen Agent Trust Hub

box-legal-workflows

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructions include a command to install the base Box MCP skill from Box's official GitHub repository (github.com/box/box-for-ai). This is a reference to a trusted vendor resource for workflow setup.
  • [INDIRECT_PROMPT_INJECTION]: The skill is intended to process external legal documents via Box AI tools, which presents a surface for indirect prompt injection. The skill mitigates this risk by explicitly defining boundaries where AI must not make decisions (e.g., legal advice, access control) and requiring human-in-the-loop confirmation for all critical actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 06:13 PM
Security Audit — agent-trust-hub — box-legal-workflows