teaching-assistant

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or data exfiltration attempts were detected in the skill files or scripts.
  • [COMMAND_EXECUTION]: The skill uses local Python scripts (quiz_generator.py, knowledge_evaluator.py, difficulty_analyzer.py) to process data. These scripts use only Python's standard library and do not perform network operations or access sensitive system files.
  • [EXTERNAL_DOWNLOADS]: No external code or packages are downloaded or executed at runtime.
  • [PROMPT_INJECTION]: The skill processes student answers for evaluation. While this creates a surface for indirect prompt injection, the instructions in SKILL.md provide clear operational boundaries to maintain the agent's educational role.
  • [CREDENTIALS_UNSAFE]: No sensitive information, such as API keys or hardcoded credentials, was found in the codebase.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 02:54 AM