spec-harness

Orchestrate spec-driven development workflows with planning, verification, and learning loops.

28|2|Updated Oct 3, 2025
One-click install
npx skills add https://github.com/chohra-med/expo_boilerplate --skill spec-harness
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spec-harness
Source: https://github.com/chohra-med/expo_boilerplate/tree/main/.claude/skills/spec-harness
Command: npx skills add https://github.com/chohra-med/expo_boilerplate --skill spec-harness

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the risk of inconsistent, low-quality feature development when using AI coding tools by enforcing a structured, validated spec-driven workflow that prevents hallucinated architecture patterns and ensures adherence to project standards.

Core Features & Use Cases

  • End-to-end workflow orchestration: Automatically routes requests through the full spec→plan→tasks→build→verify→learn flow for new feature development.
  • 3-pillar enforcement: Guarantees adherence to UAMOS (persistent memory bank), Spec-Driven Development, and the Harness ratchet/verifier/learning loop.
  • Use case: Use when building a new feature in an AI-native React Native/Expo project to generate consistent, production-ready code that matches your existing architecture without re-priming the AI tool.

Quick Start

Use the spec-harness skill to run the full end-to-end workflow for building the new user profile editing feature.

Frequently Asked Questions about spec-harness

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

FAQPage Schema
How do I prevent inconsistent code generation and hallucinated architecture patterns in AI-assisted feature development?

Spec-driven development prevents inconsistent code generation by orchestrating an end-to-end workflow through structured planning, task breakdown, implementation, and verification phases. This enforces adherence to project standards and eliminates hallucinated architecture patterns.

What is a ratchet harness for software verification in AI coding workflows?

A ratchet harness enforces software verification by applying separate verification and learning loops within AI coding workflows. It ensures production-grade output by iteratively validating generated code against structured specs and learning from failures.

How do I use spec-driven development to build new features in a React Native or Expo project?

You can use spec-driven development to build new React Native or Expo features by routing requests through a full spec, plan, tasks, build, verify, and learn flow. This generates consistent, production-ready code matching your existing architecture without re-priming.

Does AI-assisted software development require a persistent memory bank for feature development?

AI-assisted software development requires a persistent memory bank, known as UAMOS, to maintain context across feature development tasks. It guarantees adherence to the three core pillars of memory, spec-driven workflows, and verification loops.

What is the best way to orchestrate end-to-end software development workflows for AI coding tools?

The best way to orchestrate end-to-end software development workflows is by applying automated routing through structured planning, task breakdown, and iterative learning from failures. This prevents inconsistent output and ensures production-grade code generation.

Why does my AI coding tool generate code that does not match my existing project architecture?

AI coding tools generate inconsistent code without a structured, validated spec-driven workflow to enforce project standards. Applying a persistent memory bank and a ratcheted verification harness ensures generated code matches existing architecture.