autoresearch

Automate goal-directed iteration to improve measurable software metrics.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/fyodoriv/agentbrew --skill autoresearch-fyodoriv
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: autoresearch
Source: https://github.com/fyodoriv/agentbrew/tree/main/skill-plugins/dev/autoresearch
Command: npx skills add https://github.com/fyodoriv/agentbrew --skill autoresearch-fyodoriv

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Autonomous goal-directed iteration loop that drives measurable improvements in code quality and performance by systematically trying one change at a time, verifying impact, and keeping or discarding results.

Core Features & Use Cases

  • Dual-gate verification to confirm improvements without regressions.
  • Phase-based workflow (Setup, Baseline, Iteration) with Git-based history and cross-run learning.
  • Automates repetitive experimentation tasks such as reducing technical debt, increasing test coverage, or optimizing latency.

Quick Start

Start the autonomous autoresearch loop on a codebase with a defined target metric to begin iterative improvements.

Frequently Asked Questions about autoresearch

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

FAQPage Schema
How do I automate iterative code improvement for reducing technical debt?

Automating iterative code improvement requires a goal-directed loop that tries one change at a time, verifies the impact, and keeps or discards results. You need a git repository, a defined target metric, and commands for verification.

How does a dual-gate verification loop work for software metrics?

A dual-gate verification loop uses Verify and Guard gates to check each iterative change. The Verify gate confirms the target metric improved, while the Guard gate ensures no regressions occurred before keeping the change in git history.

Can I use autonomous iteration to increase test coverage in my git repository?

Yes, autonomous iteration applies to repetitive experimentation tasks like increasing test coverage. The workflow operates phase-by-phase, establishing a baseline and running automated iterations to systematically drive measurable improvements in your codebase.

What do I need to set up before starting an autonomous experimentation loop?

You need a git repository, a defined target metric to optimize, and specific commands for the Verify and Guard gates. The setup phase establishes these inputs before the baseline phase records your starting metric for comparison.

Does the autonomous iteration loop support cross-run learning and recovery?

Yes, the autonomous iteration loop supports phase-based workflow with Git-based history, logging, and recovery. Cross-run learning allows the system to use historical iteration data to improve subsequent automated changes.