joycraft-tune

Inspect CLAUDE.md, docs, and .agents/skills to score seven harness dimensions and propose upgrades.

14|1|Updated Mar 23, 2026
One-click install
npx skills add https://github.com/maksutovic/joycraft --skill joycraft-tune
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: joycraft-tune
Source: https://github.com/maksutovic/joycraft/tree/main/.agents/skills/joycraft-tune
Command: npx skills add https://github.com/maksutovic/joycraft --skill joycraft-tune

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Assesses and upgrades your project's AI development harness to enable deterministic, spec-driven progress toward Level 5, reducing ambiguity and drift in AI-driven software projects.

Core Features & Use Cases

  • Systematic 7-dimension harness assessment with actionable scoring and evidence.
  • Guided upgrade roadmap that translates findings into concrete artifacts, templates, and tasks.
  • Risk-aware, tiered upgrade workflow with traceable documentation and versioned context.

Quick Start

Run the tune assessment workflow to evaluate your project's AI development harness and plan upgrades.

Frequently Asked Questions about joycraft-tune

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

FAQPage Schema
How do I assess my AI development harness for spec quality and drift?

Assess your AI development harness by inspecting CLAUDE.md, docs, and .agents/skills to score seven dimensions including spec quality, behavioral boundaries, and testing to identify drift and propose upgrades.

What is a Level 5 AI development harness and how do I upgrade to it?

A Level 5 AI development harness enables deterministic, spec-driven progress by scoring seven dimensions like spec granularity and knowledge capture, then applying a tiered upgrade workflow with traceable documentation and artifacts.

How do I create an upgrade roadmap for my AI software factory workflow?

Create an upgrade roadmap for your AI software factory by scoring dimensions like skills, hooks, and documentation, then translating the evidence-based findings into concrete artifacts, templates, and tasks.

Can I evaluate AI project documentation and behavioral boundaries without external dependencies?

Yes, you can evaluate AI project documentation and behavioral boundaries without external dependencies by inspecting local files like CLAUDE.md and .agents/skills to generate evidence-based remedies and traceable updates.

What are the limitations of assessing AI harness maturity through manual inspection?

Manual inspection lacks the systematic seven-dimension scoring needed to reduce ambiguity, requiring a structured assessment to generate versioned context, tiered workflows, and concrete upgrade artifacts for deterministic progress.