The Agent Skills Directory

[SAFE]: The skill implements a structured pedagogical workflow for generating educational content. All operations involve local file management within the current workspace (creating directories like assets, lessons, and learning-records).
[COMMAND_EXECUTION]: The skill instructions mention using system CLI commands to open the generated HTML course files for the user. This is a standard user experience feature for local documentation agents and does not involve the execution of untrusted or arbitrary shell commands.
[PROMPT_INJECTION]: The skill incorporates external information to populate resource lists and curriculum content. While this creates a potential surface for indirect prompt injection, the instructions explicitly direct the agent to select only high-trust, high-quality resources and primary sources, which effectively mitigates the risk of ingesting malicious instructions from external data.

teach