The Agent Skills Directory

[SAFE]: The skill is designed to manage learning progress by reading and writing markdown files in a user-defined course directory (vault). These file operations are consistent with the skill's stated purpose and do not target sensitive system locations.
[PROMPT_INJECTION]: The skill processes untrusted data from user-uploaded study materials (PDFs, EPUBs, notes) to generate syllabus units and lesson content, which represents an indirect prompt injection surface. This ingestion is a primary feature of the skill's tutoring function.
Ingestion points: User-provided text and uploaded files processed during the 'Consult & anchor' stage and triggered via specific keywords.
Boundary markers: The skill instructs the agent to explicitly label its output as 'low-friction scaffolding' and to nominate 'authoritative sources' for verification, which helps contextualize the AI's generated content.
Capability inventory: The skill utilizes file-system tools to read, write, and create files and directories (e.g., index.md, syllabus.md, learner-profile.md) within the course vault.
Sanitization: No explicit code-level sanitization is defined, but the workflow focuses on generating descriptive markdown content rather than executable code.

tutor