The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute internal scripts (scripts/init_skill.py and scripts/package_skill.py) to automate skill initialization and validation. These scripts are documented as local utility tools designed to assist the developer.
[PROMPT_INJECTION]: The skill presents an indirect injection surface because its primary function is to ingest user-provided examples and descriptions to generate instructional content for new skills.
Ingestion points: User examples and queries gathered during the 'Discovery' phase described in 'references/creation-workflow.md'.
Boundary markers: The workflow for creating new skills does not explicitly require delimiters or specific instructions to ignore embedded commands in the user-provided data.
Capability inventory: The skill utilizes file system writes and local script execution to fulfill the generation process.
Sanitization: No specific sanitization or escaping logic for user-provided strings is mentioned before they are incorporated into generated SKILL.md files.
[SAFE]: The skill demonstrates safe operations by avoiding hardcoded credentials, unauthorized network requests, and sensitive file access. All external links point to the author's verified GitHub repositories.

skill-creator