The Agent Skills Directory

[Sensitive Path Access]: The skill is configured to read from a specific local configuration directory (~/.claude/plugins/config/claude-for-legal/law-student/CLAUDE.md). While this is used to personalize the learning experience, accessing files outside the immediate working directory is a pattern that should be monitored for unintended data exposure.
[Indirect Prompt Injection Surface]: The skill instructions require the agent to ingest and quote from user-provided materials such as notes, outlines, and case briefs. This creates an attack surface where instructions hidden within these external documents could potentially influence the agent's behavior.
Ingestion points: Reads local files referenced in the CLAUDE.md configuration, including user notes and outlines.
Boundary markers: The skill lacks explicit boundary markers or instructions for the agent to ignore embedded commands within the user materials it quotes.
Capability inventory: The skill primarily performs file read operations and text processing; it does not currently exhibit network or shell execution capabilities.
Sanitization: There is no evidence of automated sanitization or filtering of the content pulled from the user's law study materials.

socratic-drill