The Agent Skills Directory

[SAFE]: No executable scripts or binary files are present in the skill package; all logic is provided through instructional Markdown and JSON configuration.- [SAFE]: The skill enforces a 'manual-first' invocation posture, which ensures human oversight during the complex system design process.- [SAFE]: No evidence of obfuscation, hidden URLs, or malicious text patterns was found in any of the 11 analyzed files.- [SAFE]: The requested permissions (shell, network, and filesystem access) are not utilized by any scripts within the skill itself and are only defined in the manifest for the agent's general operational use.- [SAFE]: All external references to AI frameworks and tools (e.g., LangGraph, OpenAI SDK) are descriptive and serve a legitimate educational and architectural purpose.

agentic-development