The Agent Skills Directory

[SAFE]: The skill is a comprehensive documentation and behavioral guideline package for software design. It uses clear instructions to guide AI agents toward strategic design choices. All templates provided are static code examples using standard library modules (logging, typing, functools) with no external dependencies or dynamic execution.
[PROMPT_INJECTION]: The skill processes user-provided code for design review and refactoring, which establishes an indirect prompt injection surface. However, the skill does not utilize any dangerous tools (e.g., shell execution, network access, or sensitive file system operations) and includes explicit logic to exit silently if a task does not match a design scenario. 1. Ingestion points: Code snippets provided by the user for writing, reviewing, or refactoring tasks (SKILL.md). 2. Boundary markers: None explicitly defined within the skill instructions. 3. Capability inventory: Limited to text and code generation; complex evaluations are delegated to sibling skills which were not part of this analysis. 4. Sanitization: None.

aposd