ai-feedback-loop-optimizer

What problem does it solve?

AI outputs often require iterative refinement to reach quality levels suitable for decision-making and automation. This Skill automates the classical feedback loop (prompt → output → evaluation → improvement) to converge on the best result with minimal manual intervention.

Core Features & Use Cases

• Iterative improvement: automatically executes multiple feedback iterations
• Quality evaluation: scores outputs across multiple criteria
• Prompt optimization: tweaks prompts based on evaluation results
• Convergence判定: stops when quality meets threshold
• A/B testing: runs parallel approaches to compare strategies
• History tracking: keeps a log of iterations and outcomes
• Best output selection: selects the best performing iteration
• Parameter tuning: automatically tunes temperature/top_p
• Improvement strategies: applies multiple strategies for prompt improvement
• Early stopping: halts when improvements stall

Quick Start

Provide a task and target score; for example: "Improve a product description until you achieve 92/100 within 5 iterations." The Skill will autonomously run the feedback loop and return the best result and the improvement history.