The Agent Skills Directory

[SAFE]: The skill provides documentation and automation for skill development.
[COMMAND_EXECUTION]: The skill includes Python scripts (init_skill.py, package_skill.py, quick_validate.py) for file management tasks like initialization and packaging.
Evidence: Use of standard library functions for file system access.
Context: These scripts are intended for developer workflows and do not execute arbitrary commands or external code.
[INDIRECT_PROMPT_INJECTION]: The init_skill.py script generates content based on a user-provided skill name.
Ingestion points: The <skill-name> argument in scripts/init_skill.py used to populate the skill title and name.
Boundary markers: The generated SKILL.md template uses [TODO: ...] markers to clearly delineate areas for user modification.
Capability inventory: The skill performs file writing and directory creation via Path.mkdir and Path.write_text in scripts/init_skill.py.
Sanitization: The script uses title_case_skill_name to process input and Path.resolve() for path safety.
[PRIVILEGE_ESCALATION]: The init_skill.py script applies executable permissions to a generated example script.
Evidence: example_script.chmod(0o755) in scripts/init_skill.py.
Context: This is a standard and expected operation for making a script runnable in a new development environment.

skill-creator