The Agent Skills Directory

[SAFE]: The skill includes a dedicated 'Legal Advice Check' mechanism. This is a safety best practice that instructs the agent to identify real-world scenarios (e.g., specific names, dates, or amounts) and provide a standard refusal message directing users to professional legal resources instead of providing unauthorized legal advice.
[SAFE]: The skill accesses a specific local configuration file at ~/.claude/plugins/config/claude-for-legal/law-student/CLAUDE.md to load student context, such as learning styles and weak areas. This is a standard pattern for context-aware agents and does not involve exfiltration or unauthorized access to sensitive system directories.
[SAFE]: The instructional logic is strictly focused on pedagogical goals (Socratic questioning). It explicitly forbids providing direct answers to students, ensuring the agent remains in an educational role rather than performing tasks or generating prohibited content.

socratic-drill