harness

Orchestrate AI development lifecycle stages from requirements to implementation.

Updated Jun 6, 2026
One-click install
npx skills add https://github.com/kekemoto/skills --skill harness-kekemoto
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harness
Source: https://github.com/kekemoto/skills/tree/main/skills/harness
Command: npx skills add https://github.com/kekemoto/skills --skill harness-kekemoto

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill solves the problem of fragmented AI development by providing a centralized orchestrator that manages the lifecycle of a project, ensuring that requirements are clearly defined, reviewed, and implemented in a structured, repeatable manner.

Core Features & Use Cases

  • Workflow Orchestration: Automatically determines the current development stage (hearing, spec writing, or coding) based on repository state.
  • Safety Gates: Enforces a mandatory review process between specification and implementation to prevent premature coding.
  • Use Case: When starting a new feature, use this skill to break down ambiguous goals into actionable specifications, review them, and then delegate the implementation to the coding agent.

Quick Start

Invoke the harness skill to analyze the current repository state and determine the next step for your development goal.

Frequently Asked Questions about harness

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

FAQPage Schema
What is spec-driven AI workflow orchestration for software development?

Spec-driven AI workflow orchestration manages the AI-driven software development lifecycle by handling state transitions between requirement elicitation, specification, and implementation to ensure structured, repeatable feature delivery.

How do I automate feature branch creation and pull request generation with AI?

You can automate feature branch creation and pull request generation by invoking an orchestration skill that operates across project-specific directories to manage the complete development flow from initial specification to final PR.

How can I prevent premature coding when using an AI development workflow?

To prevent premature coding, enforce mandatory safety gates that require a formal review process between specification and implementation, validating development risks against predefined criteria before any code is written.

Does this AI orchestration workflow require any specific dependencies or environment setup?

This AI orchestration workflow operates directly within your existing project directories without requiring external dependencies, automatically analyzing repository state to determine the current development stage.

What is the best way to break down ambiguous development goals into actionable specifications?

The best way to break down ambiguous goals is using a centralized orchestrator that automatically determines the current development stage—whether hearing, spec writing, or coding—and translates requirements into actionable specifications.