harness-goal-alignment

Audit Claude Code harness components for alignment with a stated goal.

Updated Jun 30, 2026
One-click install
npx skills add https://github.com/Festo-Wampamba/Claude-Features --skill harness-goal-alignment
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harness-goal-alignment
Source: https://github.com/Festo-Wampamba/Claude-Features/tree/main/skills/harness-goal-alignment
Command: npx skills add https://github.com/Festo-Wampamba/Claude-Features --skill harness-goal-alignment

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you determine whether your Claude Code harness is actually aligned with the outcome you want, instead of accumulating conflicting rules, hooks, and skills that pull behavior in different directions.

Core Features & Use Cases

  • Full Harness Inventory: Reviews CLAUDE.md, AGENTS.md, settings, hooks, skills, agents, and plugins to map what each component does.
  • Goal Discovery and Validation: Surfaces the implied or explicit purpose of the setup and asks for confirmation when the goal is unclear.
  • Contradiction and Drift Detection: Flags components that reinforce the goal, do nothing, or actively work against it, including clashes between autonomy, safety, verbosity, and automation.
  • Use Case: Use it when your Claude Code environment feels scattered, over-restricted, or over-optimized for the wrong thing and you want a top-down audit before making changes.

Quick Start

Ask this skill to audit your Claude Code harness and report whether every component supports the single goal you actually want.

Frequently Asked Questions about harness-goal-alignment

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

FAQPage Schema
How do I audit my Claude Code harness for goal alignment and configuration contradictions?

A Claude Code harness audit reviews CLAUDE.md, AGENTS.md, hooks, skills, and agents to map component inventory, detect contradictions, and rank reinforcing versus working-against elements to expose highest-leverage changes.

Why does my Claude Code setup feel scattered and over-restricted despite clear rules?

Your Claude Code setup feels scattered or over-restricted when accumulated hooks, skills, and configuration files enforce conflicting behaviors that pull autonomy, safety, and automation in different directions, actively working against your stated goal.

What is the best way to find misalignment between CLAUDE.md and my Claude Code hooks?

The best way to find misalignment between CLAUDE.md and hooks is a top-down configuration review that maps each component's enforced behavior, classifies it as reinforcing or working-against your goal, and flags autonomy and safety clashes.

Can I use a configuration analysis to clarify the implied purpose of my Claude Code agents?

Yes, you can use configuration analysis to surface the implied or explicit purpose of your Claude Code agents and settings, prompting a goal clarification interview to confirm the outcome when the setup's true optimizing goal is unclear.

When do I need a full harness inventory for my Claude Code environment?

You need a full harness inventory when your Claude Code environment feels over-optimized for the wrong outcome, requiring a top-down audit of settings, plugins, and agents to map contradictions before making configuration changes.