The Agent Skills Directory

[SAFE]: No security issues detected. The skill defines a standard development workflow using local shell commands and project-specific tools like ai-hero-cli.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing external exercise plans to generate file structures. Ingestion points: Exercise plans parsed in the first workflow step. Boundary markers: None present. Capability inventory: The agent can create directories, write files, and execute local commands (pnpm, git). Sanitization: The skill recommends a dash-case naming convention for directories, which provides basic structural constraints.

scaffold-exercises