The Agent Skills Directory

[SAFE]: No security issues detected. The skill instructions focus on safe development practices for rapid prototyping.\n- [NO_CODE]: The skill consists entirely of markdown instructions and guidelines for the agent, with no bundled executable scripts or binaries.\n- [COMMAND_EXECUTION]: The skill encourages the creation of runnable scripts using the host project's existing task runners (e.g., pnpm, python, bun). This is the intended primary purpose of the skill and does not introduce unexpected execution vectors.\n- [DATA_EXFILTRATION]: The instructions for surfacing internal state are limited to local terminal output or UI rendering for debugging purposes and do not involve unauthorized network communication.

prototype