skills.rest
lioartoil
lioartoilCOMMUNITY
Explore all skills by this author →
GitHub

autoresearch

Optimize headless PR review prompts via iterative hill-climbing loops on GitHub.

Updated Apr 4, 2026
Search Tags:#optimization#dashboard#pr-review#score#autoresearch#headless-review#hill-climbing
One-click install
npx skills add https://github.com/lioartoil/claude-code-handbook --skill autoresearch-lioartoil
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: autoresearch
Source: https://github.com/lioartoil/claude-code-handbook/tree/main/skills/autoresearch
Command: npx skills add https://github.com/lioartoil/claude-code-handbook --skill autoresearch-lioartoil

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Hill-climbing optimizer for the headless review prompt. Runs test-score-mutate loops against a real PR to iteratively improve automated reviews and reduce manual effort in code reviews.

Core Features & Use Cases

  • Iterative optimization: runs short test-score-mutate cycles to refine the headless review prompts for PRs.
  • Real PR evaluation: operates against a live GitHub PR to ground evaluations in real diffs and comments.
  • Dashboard-driven decisions: generates scoring and changelog outputs to guide kept/reverted changes during optimization.

Quick Start

Pass a PR URL to start the hill-climbing optimization loop against a real GitHub PR.

Frequently Asked Questions about autoresearch

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

FAQPage Schema
How do I automate headless PR review optimization for GitHub repositories?▼

Automate headless PR review optimization by running an iterative hill-climbing loop that tests, scores, and mutates prompts against a live GitHub PR diff to iteratively improve automated code review quality.

What is hill-climbing optimization for automated code reviews?▼

Hill-climbing optimization for code reviews is an iterative test-score-mutate loop that refines headless review prompts by evaluating them against real PR diffs, keeping changes that improve scores and reverting those that do not.

How do I start an iterative optimization loop against a real GitHub PR?▼

Start the iterative optimization loop by passing a real GitHub PR URL along with an optimized prompt to drive the test-score-mutate cycles that evaluate and refine automated review outputs.

Can I use a scoring dashboard to guide kept and reverted decisions during PR prompt optimization?▼

Yes, you can use a scoring dashboard to guide kept and reverted decisions during PR prompt optimization by evaluating generated scoring and changelog outputs that track the effectiveness of each prompt mutation.

Do I need an optimized prompt and PR URL to run the headless review optimization cycle?▼

Yes, you need an optimized prompt, a live GitHub PR URL, and scoring dashboard integration to run the headless review optimization cycle and ground the iterative evaluations in real diffs and comments.

What are the limitations of using hill-climbing for headless PR review prompt refinement?▼

The limitation of using hill-climbing for PR review prompt refinement is that the iterative test-score-mutate loop depends entirely on the scoring dashboard's metrics to drive kept or reverted decisions, meaning poor scoring criteria yield poor prompt optimization.

Related Skills

Explore More Software Engineering Skills→
thistleknotthistleknot

autoresearch

Run autonomous hill-climbing loops to optimize scalar objectives via scored candidate changes.

Community
Intermediate
FrekiManagarmFrekiManagarm

github-code-review

Coordinate multi-agent GitHub pull request reviews across security, performance, style, and architecture.

Community
Advanced
Arize-aiArize-ai
11.0k

arize-phoenix

Capture AI application traces, evaluations, and prompts with OpenTelemetry.

Official
Advanced
xiaohei-infoxiaohei-info
28

skill-optimizer

Optimize AI agent skill definitions using an 8-dimension rubric and hill-climbing experiments.

Community
Advanced