autoresearch

Automate iterative optimization of measurable metrics with versioned configs and results tracking.

Updated Aug 27, 2026

Search Tags:#automation #workflow #metrics #experimentation #version-control #autoresearch

Ready to run

npx skills add https://github.com/VCasecnikovs/klava --skill autoresearch-vcasecnikovs

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: autoresearch
Source: https://github.com/VCasecnikovs/klava/tree/main/.claude/skills/autoresearch
Command: npx skills add https://github.com/VCasecnikovs/klava --skill autoresearch-vcasecnikovs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Autonomous optimization of any measurable metric. Edit code, measure, keep or discard, repeat. You sleep - it works. Inspired by Karpathy's autoresearch. Same core idea, generalized to anything.

Core Features & Use Cases

Autonomous loop that reads a target metric, collects the 4 required params (goal, files, metric_cmd, direction), and creates a .autoresearch.yaml config and baseline.
The loop then proposes changes to in-scope files, commits them, runs verify_cmd (if set), executes metric_cmd to extract the metric, and decides to keep or discard changes based on improvement and simplicity.
Tracks progress with a results.tsv containing the baseline and iterative experiments, enabling resumable optimization across branches.

Quick Start

Run /autoresearch to start the autonomous optimization loop against your chosen metric.

Frequently Asked Questions about autoresearch

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

FAQPage Schema

How do I automate iterative metric optimization for my codebase?▼

The autonomous optimization loop works by proposing code edits, committing them, running verification, and measuring the target metric to decide whether to keep or discard changes. It repeats this cycle automatically while tracking progress in a results.tsv file.

What is autonomous optimization and how does it apply to performance tuning?▼

Autonomous optimization is an iterative process that reads a target metric, proposes changes to in-scope files, and decides to keep or discard them based on improvement and simplicity. It applies to performance tuning by automating the edit, measure, and commit cycle.

Do I need a versioned workspace to run autonomous metric experiments?▼

Yes, a versioned workspace is required to run autonomous metric experiments. The optimization loop commits proposed changes to your files and relies on version control to track, keep, or discard iterations across branches resumably.

How do I configure automated experimentation with a YAML config file?▼

You configure automated experimentation by creating a .autoresearch.yaml file specifying four required parameters: goal, files, metric_cmd, and direction. You can also define optional verify_cmd, timeout, and max_experiments fields to control the loop.

Can I resume an autonomous optimization loop across different branches?▼

Yes, you can resume autonomous optimization across branches. The loop tracks the baseline and all iterative experiments in a results.tsv file, enabling resumable optimization so you can continue measuring progress seamlessly.

What are the limitations of using an autonomous loop for software metrics?▼

Limitations of autonomous metric optimization include the requirement for a measurable metric command and a versioned workspace. The loop relies on the defined metric_cmd to judge improvement, so unquantifiable goals or invalid verify_cmd setups will halt progress.