session

Pass

Audited by Gen Agent Trust Hub on Jul 12, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXPOSURE]: The skill reads and writes data to specific local paths: ~/.claude/plugins/config/claude-for-legal/law-student/CLAUDE.md, study-plan.yaml, and session-history.yaml. These operations are limited to the skill's local configuration directory and are used to manage study history and user preferences.
  • [COMMAND_EXECUTION]: The skill parses user input via $ARGUMENTS to determine the subject and number of questions. This is a standard functional requirement for parameter handling.
  • [REMOTE_CODE_EXECUTION]: The skill coordinates with other internal skills (bar-prep-questions, flashcards, and study-plan) to execute specialized tasks. This modular behavior uses internal agent capabilities and does not involve remote code or external script execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 12, 2026, 01:02 AM
Security Audit — agent-trust-hub — session