The Agent Skills Directory

Indirect Prompt Injection Surface: The skill is designed to ingest and analyze user-supplied case names, citations, and full case text, representing a surface for external instructions to enter the agent's context.
Ingestion points: Pasted case text and user-provided citations in SKILL.md.
Boundary markers: The skill lacks explicit delimiters or warnings to ignore embedded instructions within the pasted case text, though it focuses on verbatim extraction.
Capability inventory: The skill is limited to text generation, template creation, and reading its own configuration file; it lacks network access or the ability to execute shell commands.
Sanitization: No specific sanitization or validation logic is defined for the external case text.
Personalization Configuration: The skill accesses a configuration file at ~/.claude/plugins/config/claude-for-legal/law-student/CLAUDE.md to load user-specific formatting preferences. This is a functional requirement for providing a tailored user experience and does not target sensitive system or credential files.
Accuracy and Verification: The instructions mandate that the agent flags uncertain information and includes a clear citation check warning, mitigating risks related to AI hallucinations in a legal context.

case-brief