expert-engine

Pass

Audited by Gen Agent Trust Hub on Nov 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: No security issues detected. The skill is composed of markdown-based reasoning instructions for the AI agent and does not include any scripts, executables, or network-enabled tools.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a cognitive engine that processes user-provided problem statements, creating an attack surface for indirect prompt injection. However, because the skill itself does not define or use any tools with high-impact capabilities (such as file system access or network operations), the vulnerability is limited to the AI's internal reasoning process.
  • Ingestion points: User-supplied problem statements processed in SKILL.md (Step 0: Discovery)
  • Boundary markers: Absent
  • Capability inventory: None (the skill consists of instructions only and does not contain code or tool definitions)
  • Sanitization: Absent
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 2, 2026, 08:25 AM