socratic-tutor

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is composed of purely instructional Markdown and a YAML rubric for evaluation. There are no scripts, command execution patterns, or network requests involved.
  • [SAFE]: Data exposure risk is minimal as the skill only processes user-provided case names and readings. It does not attempt to access sensitive local files or environment variables.
  • [SAFE]: Prompt injection risks are absent; the instructions clearly define the agent's role as a supportive pedagogical tool and do not include overrides or safety bypasses.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 01:09 PM
Security Audit — agent-trust-hub — socratic-tutor