The Agent Skills Directory

[SAFE]: The skill is entirely prompt-based and does not include any executable scripts, binaries, or external code dependencies.
[SAFE]: No network access, file system operations, or tool invocations are requested or present in the skill's configuration or prompt.
[SAFE]: The system prompt is structured for educational purposes and contains no patterns indicative of prompt injection or attempts to bypass AI safety filters.
[SAFE]: While the skill ingests user-supplied data (such as task descriptions and learner profiles), the total absence of exploitable tools (e.g., shell access, file writing, or network transmission) prevents this from being an actionable indirect prompt injection vector.

differentiation-adapter