latex-econ-model

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • Prompt Injection (SAFE): No instructions attempting to override agent behavior or safety filters were detected. The instructions are purely procedural for generating LaTeX content.
  • Data Exposure & Exfiltration (SAFE): No hardcoded credentials, sensitive file paths, or unauthorized network operations were found. References are limited to reputable academic sites.
  • Unverifiable Dependencies & Remote Code Execution (SAFE): No package installations or remote script executions are present. The skill only provides text-based LaTeX code snippets for user reference.
  • Indirect Prompt Injection (SAFE): While the skill processes user input for model descriptions, it lacks any execution capabilities or tool access that could be exploited.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:08 PM