learning-study-skills

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Prompt Injection (SAFE): The skill instructions in SKILL.md and the Python code do not contain patterns intended to bypass agent safety filters or override system instructions.
  • Data Exfiltration & Exposure (SAFE): No network operations (e.g., curl, requests) or access to sensitive file paths (e.g., ~/.ssh, .env) were found. The skill only processes provided parameters.
  • Remote Code Execution (SAFE): The script does not download external code or execute remote scripts. It uses a local framework import.
  • Command Execution (SAFE): There is no usage of dangerous functions such as subprocess.run, os.system, or eval() that could lead to arbitrary command execution.
  • Code Quality (INFO): A functional bug was identified in skill.py where the variable skill_dir is referenced in the execute method but is not defined. While this will cause a runtime NameError, it does not pose a security risk.
  • Indirect Prompt Injection (SAFE): While the skill ingests untrusted data (topic, educational_level), it does not pass this data to any sensitive sinks or execute it as code. The risk is negligible.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:14 PM