evaluator-optimizer-loop
CommunityIterative critique loop for sharper outputs
Software Engineering#evaluation#quality-assurance#system-design#self-critique#iterative-improvement#red-teaming#llm-agent
Authorcropsgg
Version1.0.0
Installs0
System Documentation
What problem does it solve?
This Skill enables structured, adversarial evaluation and iterative improvement of outputs by separating an Evaluator and an Optimizer, ensuring each critique is explicit and actionable.
Core Features & Use Cases
- Separate Evaluator and Optimizer personas to provide focused critique and rewriting.
- Explicit scoring across correctness, completeness, safety, and simplicity.
- Iterative improvement with an exit condition: reach score ≥ 9 on all dimensions or stop after 3 iterations.
- Comprehensive documentation of open issues and final decisions to ship.
Quick Start
Provide an initial draft, score it across the four dimensions, then rewrite critiques and repeat until all scores meet the threshold or three iterations have occurred.
Dependency Matrix
Required Modules
None requiredComponents
Standard package💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: evaluator-optimizer-loop Download link: https://github.com/cropsgg/skills/archive/main.zip#evaluator-optimizer-loop Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 510,000+ vetted skills library on demand.