The Agent Skills Directory

[SAFE]: No security issues were identified during the analysis. The skill's behavior is entirely consistent with its stated purpose of providing a stateful tutoring experience. It manages persistent state using local markdown files (e.g., learner.md, review.md, syllabus.md) in a user-configurable directory, which is a standard pattern for agent memory. The skill also includes explicit pedagogical 'refusal rules' to ensure the AI remains in a tutoring role and does not give away answers, effectively acting as a defense against user attempts to bypass the intended learning experience. No external network connections, hardcoded credentials, or suspicious remote code patterns were found.

teach