autoresearch-agent

Automate metric-driven optimization loops by editing files, running evaluations, and recording git-backed outcomes.

Updated Apr 16, 2026
One-click install
npx skills add https://github.com/devCharuzu/philfida-taskmanage --skill autoresearch-agent-devcharuzu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: autoresearch-agent
Source: https://github.com/devCharuzu/philfida-taskmanage/tree/main/.windsurf/skills/autoresearch-agent
Command: npx skills add https://github.com/devCharuzu/philfida-taskmanage --skill autoresearch-agent-devcharuzu

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

The Autoresearch Agent automates the slow, repetitive cycle of iterating a target file to improve a measurable metric using an autonomous experiment loop.

Core Features & Use Cases

  • Automated experiment loop: the agent edits one file, runs a fixed evaluation, and commits improvements that move the metric in the right direction, while discarding unsuccessful changes via git resets.
  • Git-backed history and dashboards: tracks commits and results over time for reproducibility and comparative analysis.
  • Flexible setup: supports multiple domains, targets, evaluators, and metrics; can loop indefinitely or run single iterations as needed.

Quick Start

Create an experiment with a target file and metric, then run the single iteration to start optimizing.

Frequently Asked Questions about autoresearch-agent

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

FAQPage Schema
How do I automate metric-driven code optimization loops?

Automate metric-driven code optimization loops by configuring an agent to edit a target file, run a fixed evaluation, and commit metric improvements via git while discarding unsuccessful changes with resets.

What is an autonomous experiment loop for measurable improvement?

An autonomous experiment loop for measurable improvement is an automated workflow that iteratively edits a file, evaluates metric output, and records outcomes in a git-backed history for comparative analysis.

Do I need a git repository to run automated evaluation workflows?

Yes, you need a git repository to run automated evaluation workflows because the agent uses git commits and resets to track successful changes and discard unsuccessful iterations for reproducibility.

Can I use automated experimentation for optimizing prompts and content?

You can use automated experimentation for optimizing prompts and content because the workflow applies to any measurable improvement scenario where a reproducible git-backed target file and evaluation command exist.

What is the best way to track code optimization metrics over time?

The best way to track code optimization metrics over time is using a git-backed history that records commits and evaluation outcomes, enabling reproducibility and comparative analysis of automated experiment loops.

When should I avoid using automated metric optimization loops?

You should avoid using automated metric optimization loops when your project lacks a reproducible git-backed workflow, a specific target file, or a fixed evaluation command that outputs a measurable metric.