The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocess.run to execute standard Git commands and the GitHub CLI (gh) for repository management. These calls use structured argument lists which prevent shell injection and are limited to initializing and managing the local tutorial library in ~/coding-tutor-tutorials/.\n- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it incorporates user-provided information into its learning context.\n
Ingestion points: Data enters through the onboarding interview stored in ~/coding-tutor-tutorials/learner_profile.md and subsequent tutorial markdown files.\n
Boundary markers: Metadata is separated using YAML frontmatter and Markdown headers, providing clear structural delimitation.\n
Capability inventory: The skill possesses the ability to write to the local file system and execute Git and GitHub CLI commands via subprocess.run in setup_tutorials.py and create_tutorial.py.\n
Sanitization: The skill populates templates with user data without explicit escaping, relying on the agent's internal safety filters when the content is read back into the prompt context.

coding-tutor