HVE Core 1.1.0 Pre-release: Structured AI Workflows for VS Code

[WilliamBerryiii]

AI coding assistants excel at simple tasks. Ask for a string reversal function, and you get working code in seconds. Ask for a feature spanning twelve files across three services, and you get something that compiles cleanly and breaks everything it touches.

The problem: AI can't distinguish investigating from implementing. It writes first and verifies never.

HVE Core 1.1.0 changes that.

What "Pre-release" Means

This is not a beta. All artifacts in 1.1.0 are production-tested and marked stable. The pre-release designation gives early adopters access to the VS Code extension's pre-release channel, which includes:

Channel	What You Get
Stable	Production-ready artifacts only
Pre-release	Stable + preview + experimental artifacts

Switch channels anytime from the extension page. New artifacts start as experimental, graduate to preview when feature-complete, then reach stable after production validation.

VS Code Extension

Install from the Marketplace with zero configuration:

Prefer customization? The hve-core-installer agent offers seven installation methods: multi-root workspace, submodule, peer clone, git-ignored, mounted directory, GitHub Codespaces, or extension-only.

RPI Workflow

Research → Plan → Implement. Three phases that transform how AI approaches complex tasks.

When AI knows it cannot implement during research, it stops optimizing for "plausible code" and starts optimizing for "verified truth." The constraint changes the goal.

New in 1.1.0:

• Complete RPI documentation explaining the psychology behind constraint-based workflows
• 15-minute first workflow tutorial
• rpi-agent for autonomous mode when strict phase separation isn't needed

18 Specialized Agents

Category	Agents
Development Workflow	rpi-agent, task-researcher, task-planner, task-implementor, pr-review, github-issue-manager
Architecture & Docs	adr-creation, arch-diagram-builder, brd-builder, prd-builder, prompt-builder, security-plan-creator
Azure DevOps	ado-prd-to-wit
Data Science	gen-data-spec, gen-jupyter-notebook, gen-streamlit-dashboard, test-streamlit-dashboard
Setup	hve-core-installer

MCP Server Integration

Agents now include guidance for configuring MCP (Model Context Protocol) servers. Connect to Azure DevOps, GitHub, Microsoft Docs, and Context7 for richer context during research and planning phases.

Security and Quality

• OpenSSF Scorecard compliance: branch protection, pinned dependencies, minimal permissions
• CodeQL security analysis on every PR
• Pester test infrastructure: 250+ PowerShell tests with reusable CI workflow
• Checksum validation for all tool downloads

Get Started

1. Install the VS Code extension
2. Complete the first workflow tutorial (15 minutes)
3. Explore RPI documentation

We Want Your Feedback

What works? What doesn't? What should we build next?

• Report issues
• Start a discussion
• Contribute