Getting Started

Repo Hub gives your AI coding agent everything it needs to ship real features — context across your codebase, structured agent workflows, and infrastructure access. Follow these steps to set up your first workspace.

Prerequisites

• Node.js 18+ installed
• Git configured with SSH access to your repositories
• Cursor, Kiro, OpenCode, or any AI-powered editor installed

Quick Start

Every command can be run with npx @arvoretech/hub <command>. To use the shorter hub alias throughout this guide (and all other docs), install it globally first:

npm i -g @arvoretech/hub

1. Initialize a new hub

hub init my-hub
cd my-hub

This launches an interactive TUI that walks you through naming your workspace, choosing your AI editor, adding repositories, selecting agents and skills from the directory registry, picking MCP servers, and choosing your config format (YAML or TypeScript).

At the end, you’ll have a fully configured workspace with your chosen config file (hub.yaml or hub.config.ts), agents, skills, and a .gitignore. See Configuration for details on both formats.

2. (Optional) Add more repositories

If you didn’t add all repos during init, you can add them later:

hub add-repo git@github.com:company/api.git --tech nestjs
hub add-repo git@github.com:company/frontend.git --tech nextjs

Each repository is declared in your config file with its tech stack, commands, and environment configuration.

3. Set up the workspace

hub setup

This clones all repos, starts infrastructure services (databases, caches), and installs dependencies.

4. Generate editor configuration

hub generate --editor cursor      # or
hub generate --editor kiro         # or
hub generate --editor claude-code  # or
hub generate --editor opencode

This reads your config file (hub.yaml or hub.config.ts) and produces editor-specific files:

• Cursor: .cursor/rules/orchestrator.mdc, .cursor/rules/<steering>.mdc, .cursor/agents/*.md, .cursor/skills/hub-docs/, .cursor/mcp.json, .gitignore + .cursorignore
• Kiro: .kiro/steering/orchestrator.md, .kiro/steering/<steering>.md, .kiro/agents/, .kiro/skills/hub-docs/, .kiro/settings/mcp.json, AGENTS.md, <name>.code-workspace, .gitignore
• Claude Code: CLAUDE.md (with steering appended), .claude/agents/, .claude/skills/hub-docs/, .mcp.json, .gitignore
• OpenCode: opencode.json, .opencode/rules/orchestrator.md, .opencode/rules/<steering>.md, .opencode/agents/, .opencode/skills/hub-docs/, .opencode/commands/, .opencode/plugins/, .gitignore

All editors also get a hub-docs skill automatically fetched from hub.arvore.com.br, giving agents access to Repo Hub documentation. Steering files from the steering/ directory are distributed to each editor in its native format.

5. Open and start building

Open the project in your editor. Your AI now sees all repos and follows the generated workflow.

How the Context Pattern Works

The magic is in two lines of config:

# .gitignore — repos are excluded from the hub's git
api
frontend

# .cursorignore — but included for AI context
!api/
!frontend/

Your AI sees all repos as one workspace. Each repo keeps its own git history. Zero migration, zero overhead.

What Happens When You Generate

When you run hub generate --editor cursor, the CLI:

1. Reads your config file (hub.yaml or hub.config.ts) declaring the full pipeline
2. Generates an orchestrator rule with workflow instructions
3. Copies steering files from steering/ to the editor’s native format
4. Creates agent definitions for each pipeline step
5. Fetches the hub-docs skill from hub.arvore.com.br
6. Configures MCP connections for infrastructure access
7. Sets up .gitignore and .cursorignore for the context pattern

The editor is the runtime. There is no daemon, no server, no separate process. Your AI editor executes the workflow by following the generated rules.

Next Steps

• Configuration — Learn the full config schema (YAML and TypeScript)
• Agent Orchestration — Understand how agents collaborate
• Skills — Package domain knowledge for agents
• MCPs — Connect AI to your infrastructure