The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes Python and TypeScript scripts (init_skill.py, init_skill.ts) that automate the creation of directory structures and files on the local file system.
[COMMAND_EXECUTION]: Initialization scripts use chmod +x to ensure that generated example scripts in the scripts/ directory are executable by the user.
[EXTERNAL_DOWNLOADS]: The init_skill.ts script uses npx to execute tsx, which may involve downloading the tsx package from the official npm registry if it is not present in the local environment.
[PROMPT_INJECTION]: The skill processes user-provided inputs (descriptions and workflow examples) to generate content for new skills, creating an indirect prompt injection surface.
Ingestion points: User input provided during the skill planning and editing steps described in SKILL.md.
Boundary markers: Validation scripts check for specific characters (e.g., angle brackets) in descriptions to provide a basic layer of separation.
Capability inventory: The skill can create directories and write files to the local system via its initialization scripts.
Sanitization: Basic validation of the skill name (hyphen-case) and description (no angle brackets) is implemented in the quick_validate scripts.

skill-creator