The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill contains a reference to a configuration file hosted on a repository belonging to a trusted organization. This reference is used for documentation and alignment with established standards.
[PROMPT_INJECTION]: The skill processes code provided in the agent's context (Ingestion point: user-provided code). It does not specify explicit boundary markers or input sanitization. However, the skill does not utilize any dangerous tools or scripts (Capability inventory: none), and the instructions focus strictly on logic-preserving refactoring. The author-provided metadata 'risk: critical' was reviewed and found to be non-malicious in this context.

code-simplifier