autoresearch-agent

Run autonomous experiment loops to optimize code by a measurable metric.

Updated Apr 2, 2026

Search Tags:#performance improvement #code optimization #autonomous testing #experiment loop #autoresearch

Ready to run

npx skills add https://github.com/4lerman/text_evaluator --skill autoresearch-agent-4lerman

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: autoresearch-agent
Source: https://github.com/4lerman/text_evaluator/tree/main/.agents/skills/engineering-advanced-skills/autoresearch-agent
Command: npx skills add https://github.com/4lerman/text_evaluator --skill autoresearch-agent-4lerman

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the process of experimenting and optimizing code, allowing users to run autonomous loops that optimize any file by a measurable metric.

Core Features & Use Cases

Automated Experiment Loops: Run experiments that optimize code by a measurable metric, without manual intervention.
File Editing and Evaluation: Automatically edit target files, commit changes, evaluate the results, and keep improvements or discard failures.
Use Case: Use this Skill to optimize the performance of a code function, reducing execution time or memory usage. It can be used for any measurable improvement loop, such as optimizing code speed, reducing bundle/image size, improving test pass rate, optimizing prompts, or improving content quality.

Quick Start

Set up the experiment by running: /ar:setup engineering/api-speed --target src/api/search.py --eval "pytest bench.py" --metric p50_ms --direction lower

Frequently Asked Questions about autoresearch-agent

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I automate code optimization for a specific file?▼

You can automate code optimization by running autonomous experiment loops that incrementally edit target files, evaluate the results, and keep improvements or discard failures. This requires setup with a target file, evaluation command, and metric definition.

Can I use autonomous testing to reduce execution time and memory usage?▼

Yes, autonomous testing via experiment loops applies to any measurable code improvement scenario, including reducing execution time, lowering memory usage, fixing bugs, and improving test pass rates.

What do I need to set up an automated experiment loop for performance improvement?▼

To set up an experiment loop for performance improvement, you need to define a target file, an evaluation command to measure results, and a specific metric with a direction to optimize.

Does autoresearch work for optimizing non-functional metrics like bundle size?▼

Autoresearch works for any measurable improvement loop, including optimizing code speed, reducing bundle or image size, improving test pass rates, optimizing prompts, and improving content quality.

What's the best way to run experiments that incrementally edit and evaluate code?▼

The best way to run incremental code experiments is using an autonomous loop that automatically edits files, commits changes, evaluates results against a metric, and keeps improvements while discarding failures.

When should I not use autonomous experiment loops for code optimization?▼

You should not use autonomous experiment loops when a code improvement scenario lacks a clear, measurable metric or when you cannot provide an evaluation command to assess the results.