autoresearch

1# Autoresearch: Autonomous Iterative Experimentation
2 
3An autonomous experimentation loop for any programming task. You define the goal and how to measure it; the agent iterates autonomously -- modifying code, running experiments, measuring results, and keeping or discarding changes -- until interrupted.
4 
5This skill is inspired by [Karpathy's autoresearch](https://github.com/karpathy/autoresearch), generalized from ML training to **any programming task with a measurable outcome**.
6 
7---
8 
9## Agent Behavior Rules
10 
111. **DO** guide the user through the Setup phase interactively before starting the loop.
122. **DO** establish a baseline measurement before making any changes.
133. **DO** commit every experiment attempt before running it (so it can be reverted cleanly).
144. **DO** keep a results log (TSV) tracking every experiment.
155. **DO** revert changes that do not improve the metric (git reset to last known good).
166. **DO** run autonomously once the loop starts -- never pause to ask "should I continue?".
177. **DO NOT** modify files the user marked as out-of-scope.
188. **DO NOT** skip the measurement step -- every experiment must be measured.
199. **DO NOT** keep changes that regress the metric unless the user explicitly allowed trade-offs.
2010. **DO NOT** install new dependencies or make environment changes unless the user approved it.
21 
22---
23 
24## Phase 1: Setup (Interactive)
25 
26Before any experimentation begins, work with the user to establish these parameters.
27…