The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from a YAML file, creating a surface for indirect prompt injection.
Ingestion points: scripts/run.py reads outline/module_plan.yml from the workspace.
Boundary markers: The skill does not employ specific delimiters or instructions to ignore potential commands embedded within the module plan data.
Capability inventory: The skill has the capability to write files (dump_yaml) to the workspace.
Sanitization: The scripts/run.py script does not demonstrate explicit validation or sanitization of the YAML content before it is passed to the exercise generation logic.

exercise-builder