Skip to content

Latest commit

 

History

History

README.md

title HVE Core Documentation
description Documentation hub for HVE Core, a prompt engineering framework that brings AI-powered agents, prompts, instructions, and skills to your GitHub Copilot workflow
sidebar_position 1
author Microsoft
ms.date 2026-07-15
ms.topic overview
keywords
hve core
documentation
copilot customizations
agents
prompt engineering
estimated_reading_time 3

HVE Core gives your team production-ready agents, reusable prompts, coding instructions, and executable skills for GitHub Copilot. You get structured workflows (Research → Plan → Implement), schema-enforced quality gates, and role-specific tooling across 10 engineering disciplines. Install from the VS Code Marketplace and start shipping with AI-assisted engineering in minutes.

Choose Your Extension

Option HVE Core All HVE Installer
What you get Every agent, prompt, instruction, and skill in the framework Pick only the collections you need
Best for Teams that want the full toolkit out of the box Teams that prefer a curated, lightweight setup
Install Install HVE Core All Install HVE Installer

Not sure which to choose? See the installation methods comparison for a detailed breakdown.

Find Your Path

New to HVE Core?

Get up and running with installation, configuration, and your first AI-assisted workflow.

Leading a Team?

Set up HVE Core for your team with governance, collections, and customization options.

Contributing to HVE Core?

Create and maintain agents, prompts, instructions, and skills for the framework.

Going Deeper?

Explore advanced capabilities including Design Thinking coaching, security planning, and methodology reference.

Roles

HVE Core provides dedicated tooling for 10 engineering roles, each with curated agents, prompts, and starter workflows. Find your role guide on the Role Guides page.

AI-Assisted Project Lifecycle

HVE Core supports a 9-stage lifecycle from initial setup through ongoing operations. Each stage maps to specific agents, prompts, and role-specific guidance.

Explore the full lifecycle →

Agent Systems

Specialized agents are organized into functional groups that combine agents, prompts, and instruction files into cohesive workflows.

  • RPI Orchestration separates complex tasks into research, planning, implementation, and review phases
  • Project Planning creates ADRs, BRDs, PRDs, architecture diagrams, and security plans through guided AI workflows
  • GitHub Backlog Manager automates issue discovery, triage, sprint planning, and execution
  • Additional systems are documented in the Agent Catalog

Browse the Agent Catalog →

RPI Methodology

Research, Plan, Implement, Review (RPI) decomposes complex engineering tasks into phase skills coordinated by RPI Agent or invoked directly.

  • Why RPI? explains the problem statement and design rationale
  • RPI overview introduces RPI Agent, /rpi, and the direct rpi-* phase skills
  • Using Together describes phase coordination and durable handoffs

RPI Documentation →

Design Thinking

The dt-coach agent guides teams through nine Design Thinking methods across problem space, solution space, and validation.

Browse all Design Thinking docs →

Prompt Engineering

HVE Core structures AI artifacts with protocol patterns, input variables, and evidence-backed quality gates.

  • The hve-builder skill creates, improves, refactors, replaces, reviews, and validates AI artifacts through one lifecycle
  • AI Artifacts Overview covers common patterns across artifact types
  • Activation Context explains when artifacts activate within workflows

Quick Links

Resource Description
Customization Guide Adapt HVE Core to your workflow
CHANGELOG Release history and version notes
CONTRIBUTING Repository contribution guidelines
Scripts README Automation script reference
Extension README VS Code extension documentation

🤖 Crafted with precision by ✨Copilot following brilliant human instruction, then carefully refined by our team of discerning human reviewers.