eric-sabe

Engsys — AI skill for Claude Code

AI community

An AI engineering team you install into any Claude Code project — stack-agnostic agents plus a deterministic installer.

How to install Engsys

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

What Engsys does

An AI engineering team you install into any Claude Code project — stack-agnostic agents plus a deterministic installer. Swap stack packs, not prompts.

Alternatives in AI

Gem Team — Turn AI coding into an engineering process 210 ★
Llmix — Production LLM call layer for AI agents and tools: keep OpenAI/Anthropic/AI SDK/LiteLLM, hot-swap models with 132 ★
Coding Agent Account Manager — Sub-100ms auth switching for AI coding CLIs (Claude Code, Codex, Gemini): swap subscription accounts instantly 111 ★

README

engsys

**An AI engineering team you install into any project.**

A complete Claude Code engineering system — agent personas, slash commands, skills, hooks, workflow docs, and curated lessons — plus a deterministic installer that materializes it **faithfully** into any project from one config file.

**[Live explainer ↗](https://eric-sabe.github.io/engsys/)** — or open [`index.html`](index.html) locally — for a visual tour of the system and the team.

Why

The system used to live copy-pasted across projects at different maturity levels. That caused two recurring pains:

Re-naturalization tax — hand-editing agent profiles per stack (the cloud architect as AWS vs Azure vs GCP) on every import.
Unfaithful plumbing — when commands were ported and a model was asked to "set up the folders," it improvised and skipped steps.

engsys fixes both. See [`docs/architecture.md`](docs/architecture.md) for the full design.

The core idea: three layers

Layer	Question	Stability	Lives in
Persona	Who does the work	Stable everywhere	`core/agents/`
Capability	Which stack/tech	Chosen per project	`stacks/**/skills/` (packs)
Project facts	What's true about this repo	Unique per project	generated `CLAUDE.md`

The cloud architect (Melvin) never changes; you install `cloud-architecture-aws` **or** `-azure` **or** `-gcp` **or** `-cloudflare` and he auto-loads whichever is present. Adapting to a project becomes *choosing packs*, not *editing prose*.

Quickstart

**Option A — install from npm** (global CLI):

npm install -g engsys

cd your-project
engsys init                 # scaffold engsys.config.yaml from the bundled example
$EDITOR engsys.config.yaml  # pick cloud / iac / lang / platform / db / agents
engsys install --into .     # materialize .claude/, CLAUDE.md, settings, .mcp.json

# open the project in Claude Code and run /naturalize  (the one model-d