harness-init

Automate Claude Code harness setup with rules, hooks, memory, and routing.

123|22|Updated Apr 7, 2026
One-click install
npx skills add https://github.com/AlexZio00/claude-code-skills --skill harness-init-alexzio00
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harness-init
Source: https://github.com/AlexZio00/claude-code-skills/tree/main/harness-init
Command: npx skills add https://github.com/AlexZio00/claude-code-skills --skill harness-init-alexzio00

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates the setup of Claude Code harness infrastructure, delivering substantive rules, hooks, memory models, and routing definitions. It ensures domain presets provide real content, not skeletons, and enforces violation testing and reject-by-default discipline. It is not for project scaffolding; use project-init for that.

Core Features & Use Cases

  • Generates hard rules with real content (Tier 0) from domain presets, including violation testing.
  • Creates and merges harness wiring: rules, hooks, memory, and routing definitions, while preserving existing configurations.
  • Supports interactive domain selection, memory strategy, and optional agent routing for orchestrated AI workflows.

Quick Start

Provide the harness configuration request by saying '/harness-init' to generate rules, hooks, memory, and routing.

Frequently Asked Questions about harness-init

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

FAQPage Schema
How do I set up rules, memory, and hooks for Claude Code?

Harness initialization generates hard rules with real domain content from presets, structured memory with handoff capabilities, and routing definitions to support agent orchestration and pipeline review gates.

How do I configure domain presets and guardrails for AI agent workflows?

You configure guardrails by selecting interactive domain presets during setup, which generates Tier 0 hard rules with real content, violation testing, and structured memory persistence for your project.

Can I merge new hooks and memory configurations without overwriting existing ones?

The setup process merges new hooks and memory configurations into your project while preserving existing configurations, ensuring merge-safe updates to your Claude Code harness infrastructure.

What is the difference between harness setup and project scaffolding?

Harness setup configures Claude Code infrastructure including rules, memory, hooks, and routing for guardrails and pipeline review gates, while project scaffolding handles initial project structure and requires a separate project-init tool.

When should I not use automated harness configuration for my project?

Avoid harness setup if your project only needs basic scaffolding without guardrails, or if it does not require memory persistence, pipeline review gates, or agent orchestration infrastructure.