Last commit on May 26, 2026·Created on March 22, 2026
krzysztofdudek/ResearcherSkill
“An autonomous loop for optimizing measurable metrics through repeated experimentation”
Combined rank
#644
across all skills
Stars
258
7d change unavailable
Forks
27
Watchers
2
Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of real attention, adoption, and active maintenance.
TL;DR
Researcher automates the iterative process of tuning configurations and improving performance metrics. It is designed for tasks where a goal—such as reducing latency, increasing accuracy, or improving build times—can be measured quantitatively or scored against a rubric.
The skill operates independently, managing its own experimental branches and logging results to a dedicated lab directory. It continues to iterate until a target threshold is met or a performance plateau is reached, removing the need for manual trial-and-error.
WHO IT'S FOR
Performance / Infrastructure Engineers
optimizing build times and system latency
Prompt Engineers
tuning prompt quality against a rubric
ML / Data Scientists
finding best configurations for model accuracy
QA / Test Automation Engineers
improving test pass rates autonomously
Want to install it?
Claude Marketplace · Claude Plugin · Clone to Claude · Manual Copy · Symlink to AgentRepository instructions