The Agent Skills Directory

[SAFE]: The skill performs expected project initialization tasks, including directory creation and file updates within a defined workspace (projects/{slug}/). No external network requests or suspicious command executions were found.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied slug and intent strings which are written to README.md, index.md, and log.md.
Ingestion points: Inputs section defines slug and intent as user-provided data.
Boundary markers: None explicitly defined in the templates or instructions.
Capability inventory: File system write operations (mkdir, template rendering, file appending).
Sanitization: The instructions require the slug to be lowercase and hyphenated, providing basic structural validation. This is a standard risk profile for documentation-focused skills.

proj-init