yuri-semenenko

AI Engineering Workspace — AI skill for Claude Code

AI community

One engineering workflow across Claude Code, Codex, Copilot, and Gemini CLI.

How to install AI Engineering Workspace

This entry records only its repository, not the path inside it, so there is no exact command to give. Open yuri-semenenko/ai-engineering-workspace and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What AI Engineering Workspace does

One engineering workflow across Claude Code, Codex, Copilot, and Gemini CLI. A portable persona canon, process skills, and safety guardrails, kept in sync by design.

Alternatives in AI

Skills Manager — A lightweight desktop app to manage, sync, and organize AI agent skills across 15+ coding tools — Cursor, Clau 813 ★
Skillfish — The skill manager for AI coding agents 312 ★
Gem Team — Turn AI coding into an engineering process 210 ★

README

AI Engineering Workspace

**One engineering workflow. Every AI coding assistant.**

[![CI](https://github.com/yuri-semenenko/ai-engineering-workspace/actions/workflows/ci.yml/badge.svg)](https://github.com/yuri-semenenko/ai-engineering-workspace/actions/workflows/ci.yml) [![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](./LICENSE)

A personalized AI engineering workspace for building one consistent engineering workflow across Claude Code, OpenAI Codex, GitHub Copilot, and Gemini CLI.

This is not a prompt pack and not a codegen shortcut, and it is not about building AI systems. It is a portable methodology layer for *using* AI to do software engineering: a fixed engineering canon, a personalized identity layer, validated skills/commands/prompts, and tool-specific adapters.

AI tools change. Engineering principles don't. Encode the principles once and let them outlive the tools.

What this gives you

One engineering methodology across every supported AI tool.
A persona wizard that adapts the assistant to your discipline, seniority, workflow, stack, and tooling.
A generated recommended-skills view that tells you which shipped skills to reach for first.
A /start onboarding entrypoint inside each supported tool.
A tool-agnostic repository-contract convention (AGENTS.md) and a workflow that writes one from a repository's own evidence.
Drift guards and CI checks so templates, mirrors, and skill references do not silently rot.
ADRs that document why the architecture works this way.

Why

Enterprise engineers rarely get to pick their AI assistant, and they often use more than one. Each tool wants its configuration in a different place and format, so the usual result is several divergent piles of per-tool prompts that drift apart the moment you edit one. That is a dotfiles problem dressed up as engineering.

This kit inverts it. The durable layer — how you make decisions, review code, write an RFC, define "done" — is w