config-tune

Analyzes ANALYSIS.md outputs to tune Claude Code configuration files with dry-run option and rubric-based evaluation.

Updated Jan 13, 2026
One-click install
npx skills add https://github.com/yoshihiko555/dotfiles --skill config-tune
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: config-tune
Source: https://github.com/yoshihiko555/dotfiles/tree/main/shared/skills/common/config-tune
Command: npx skills add https://github.com/yoshihiko555/dotfiles --skill config-tune

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Analyzes ANALYSIS.md to interactively tune Claude Code configuration files, ensuring alignment with Skill/Agent/CLAUDE.md/Rule and rubric requirements.

Core Features & Use Cases

  • Phase-driven, interactive tuning based on ANALYSIS.md outputs
  • Dry-run support and per-item approvals to prevent unintended edits
  • Re-generation of ANALYSIS.md after changes and references to rubric for evaluation

Quick Start

Provide the target configuration file path to start an interactive tuning session.

Frequently Asked Questions about config-tune

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

FAQPage Schema
How do I tune Claude Code configuration files based on an analysis document?

You can tune Claude Code configuration files by analyzing ANALYSIS.md outputs interactively, aligning edits with CLAUDE.md and Rule files to ensure configurations meet rubric requirements.

Can I preview configuration changes before applying them to my Claude Code project?

Yes, interactive config tuning supports a dry-run option and per-item approvals, allowing you to review and selectively apply configuration edits to prevent unintended changes.

What is the best way to automate repeatable config tuning across multiple Claude Code environments?

Automated config tuning uses a phase-driven workflow to apply repeatable edits across multiple files and environments, evaluating results against a rubric to enforce consistency.

How does rubric-based evaluation work for Claude Code config edits?

Rubric-based evaluation assesses applied configuration edits against criteria defined in references, ensuring tuned files comply with alignment rules and generating a post-change ANALYSIS.md for verification.

Do I need an existing ANALYSIS.md file to start tuning Claude Code configurations?

Yes, an existing ANALYSIS.md is required as input to map the tuning phases, and the tool regenerates this analysis after changes to verify alignment with the updated configuration.

Are there limitations to interactive config tuning for Claude Code projects?

Config tuning is limited to projects using Claude Code configuration structures, requiring explicit file paths and per-item approval to safely manage edits across multiple targeted environments.