The Agent Skills Directory

[SAFE]: No security issues or malicious patterns were detected. The skill serves as a template for developers to explore design options through temporary code.
[COMMAND_EXECUTION]: The instructions guide the agent to add run scripts to standard project configuration files like package.json, Makefile, or pyproject.toml. These commands use existing local runtimes (e.g., pnpm, python, bun) to execute the created prototype code.
[DATA_EXPOSURE]: The skill explicitly mandates a "no persistence by default" policy. It recommends using in-memory storage or clearly marked local scratch files for state, which prevents sensitive data from being accidentally committed or persisted in production environments.

prototype