The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill reads and processes data from user-editable files in the local workspace to steer its teaching logic, which creates a surface for indirect prompt injection findings.
Ingestion points: The agent ingests data from MISSION.md, RESOURCES.md, NOTES.md, and files within the ./learning-records/ directory.
Boundary markers: Absent; the skill lacks specific delimiters or warnings to ignore instructions that might be embedded in the ingested Markdown content.
Capability inventory: The skill has the ability to write to the file system (creating lessons and assets) and execute CLI commands to open files.
Sanitization: The instructions do not specify any validation or sanitization for the content read from external workspace files.
[DYNAMIC_EXECUTION]: The skill is instructed to author and reuse components in the ./assets/ directory, including "quiz widgets" and "simulators," which involves the generation and local storage of executable script content.
[COMMAND_EXECUTION]: The instructions suggest that the agent should run a CLI command to open generated lesson files for the user, involving local process invocation.

teach