optimization-controller
CommunityIteratively improve agent performance.
Software Engineering#performance tuning#experimentation#agent optimization#iterative improvement#langfuse#evaluation loop
Authormberto10
Version1.0.0
Installs0
System Documentation
What problem does it solve?
This Skill provides a structured, automated process for iteratively improving the performance of AI agents by systematically identifying, hypothesizing, experimenting with, and analyzing changes to their configuration or prompts.
Core Features & Use Cases
- 5-Phase Execution Loop: Guides the optimization process through Diagnose, Hypothesize, Experiment, Analyze, and Compound phases.
- Lever Tuning: Manages the application of changes to specific agent parameters (levers) in single or multi-lever modes.
- Strict Evaluation: Enforces preflight checks and strict guardrails to prevent regressions.
- Use Case: An agent is underperforming on a specific metric. This Skill can automatically test different prompt variations or configuration settings, evaluate their impact, and decide whether to keep or rollback the changes, all within a controlled loop.
Quick Start
Use the optimization-controller skill to begin optimizing the 'article-writer' agent.
Dependency Matrix
Required Modules
pyyamllangfuse
Components
scriptsreferences
💻 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: optimization-controller Download link: https://github.com/mberto10/mberto-compound/archive/main.zip#optimization-controller 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 471,000+ vetted skills library on demand.