The Agent Skills Directory

[SAFE]: The skill serves as an orchestration framework for delegating implementation work. It provides structured guidance on how to frame tasks for external coding agents (like Codex or Claude Code) without introducing any automated execution or sensitive data access within its own instructions.
[SAFE]: The skill uses markdown and JSON for documentation and metadata. No scripts, binaries, or remote code patterns are included. All dependencies are internal references to other documentation files within the skill package.
[SAFE]: Analysis of the provided prompts and examples shows a focus on software engineering best practices, such as defining acceptance criteria, enforcing validation (tests/linting), and maintaining repository-aware context. No prompt injection or obfuscation techniques were identified.

coding-agent