# Autoresearch · pjhoberman

> Autonomous experiment loops on any codebase — one file, one metric, one loop. An agent edits a constrained file, runs an eval, keeps improvements, reverts failures, and repeats unattended. Based on Karpathy’s autoresearch pattern, generalized beyond ML training to any code with a measurable outcome.

## Facts
- Page: https://tashan.sh/capability/plugin-pjhoberman-autoresearch-autoresearch
- tashan id: plugin:pjhoberman/autoresearch/autoresearch
- Source: https://github.com/pjhoberman/autoresearch
- Type: plugin
- Category: other
- tashan score: 29.0 / 100
- Adoption: 24.0
- Upkeep: 43.0
- Freshness: 53.0
- Evidence coverage: 84% of the inputs this score can use
- Health: active
- Instruction depth: solid
- GitHub stars: 21
- Official: no

## Install

```sh
/plugin marketplace add anthropics/claude-plugins-community
/plugin install autoresearch@claude-community
```

## Security audit
Not scanned. We audit npm-published capabilities; this one has no npm package we can resolve, or has not reached the queue. This is not a clean bill of health.

---
Measured 2026-09-13 by tashan (https://tashan.sh) from public evidence. Scorer s5.