The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes local utility scripts, specifically scripts/init_skill.py, scripts/generate_openai_yaml.py, and scripts/quick_validate.py, for the purpose of initializing skill directories, managing metadata, and validating structural integrity. These tools are used within a local development environment and are appropriate for the skill's stated mission.
[COMMAND_EXECUTION]: The development workflow described in the skill includes instructions for the agent to execute and test newly created scripts in the scripts/ directory to ensure deterministic results and reliability.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it ingests user-provided functional examples and descriptions to generate instructional content for other skills.
Ingestion points: User-provided use cases and concrete examples collected during the skill design phase.
Boundary markers: While the skill encourages structured YAML frontmatter for metadata, it does not explicitly mandate delimiters or 'ignore' instructions for content derived from user examples.
Capability inventory: The skill framework allows for file writing and the execution of local shell scripts.
Sanitization: No explicit sanitization or validation protocols are mentioned for user-supplied data that is incorporated into the generated SKILL.md or supporting scripts.

skill-creator