Last commit on October 31, 2026·Created on March 13, 2026
uditgoenka/autoresearch
“Improve a measurable goal through automated experiments.”
Combined rank
#94
across all skills
Stars
6.8k
+0.5% in last 7d
Forks
493
-0.2% in last 7d
Watchers
24
0.0% in last 7d
Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
TL;DR
Autoresearch provides a structured framework for AI agents to perform autonomous, goal-directed iteration. It implements a modify-verify-discard cycle that allows an agent to refine its output against specific metrics without constant manual intervention.
This skill is designed for automation builders and developers using Claude Code to handle repetitive optimization tasks. By automating the verification process, it transforms open-ended research into a systematic loop of improvement.
WHO IT'S FOR
AI agent / automation builders
implementing autonomous goal-directed iteration loops
Claude Code power users
automating repetitive code optimization tasks
ML / AI researchers
optimizing models against specific performance metrics
Technical leads / architects
designing self-healing or self-optimizing systems
Categories
Claude
Codex
OpenCodeWant to install it?
Claude Marketplace · Claude Plugin · Clone to Claude · Clone to Codex · Manual Copy · NPX Skills AddRepository instructions