implement-agent-reflection

What problem does it solve?

This Skill enables agents to design self-correcting loops for LLM-based tasks, ensuring high-quality outputs by forcing critical self-reflection before finalizing the result.

Core Features & Use Cases

• Self-Correcting Loops: Establishes an Evaluator-Generator pattern that critiques initial outputs and iteratively refines them.
• High-Quality Outputs: Prevents subtle flaws in single-shot LLM outputs through structured reflection.
• Use Case: Ideal for scenarios like code review or prompt engineering, where ensuring a high standard of output is critical.

Quick Start

Use the implement-agent-reflection skill to refine an output: trigger-reflection-on-output <your-output>.