Gorig: AI-Native Backend Delivery System

English | 简体中文

Gorig is a backend delivery system for the AI era. It combines a Go web service framework, gorig-cli, and the gorig-backend AI skill so teams can turn product intent into structured, verified, and operable backend services.

📚 Project Wiki: https://deepwiki.com/jom-io/gorig
🔧 Operations Dashboard: https://github.com/jom-io/gorig-om

Why Gorig

AI can generate backend code quickly, but speed alone is not enough. Real teams still need stable architecture, source-aware framework usage, tests, documentation, runtime visibility, and a path from local development to production operation.

Gorig packages those needs into one delivery workflow:

Product value	What teams get
Ship from intent	Describe a business need such as CRUD, login, reminders, live updates, async jobs, or deployment prep, then generate a runnable Go service shape.
Keep AI on rails	Preserve consistent Router -> Controller -> Service -> Model boundaries instead of accepting one-off generated code.
Verify before trust	Treat AI output as a delivery artifact with build checks, tests, smoke verification, and generated docs.
Operate after launch	Use built-in patterns for health, logs, scheduling, messaging, SSE, auth, configuration, graceful shutdown, and rollback planning.
Grow as an ecosystem	Add gorig-node, gorig-hub, and gorig-om when services need discovery, SDK generation, runtime nodes, and operations visibility.

Delivery Workflow

Gorig is designed around the full backend lifecycle, not a single scaffolding command.

Step	What happens
1. Describe	Start from product language: customer management, order workflow, reminder task, admin API, service deployment.
2. Scaffold	gorig-cli creates projects, modules, CRUD services, tests, docs, and environment configuration.
3. Implement	The gorig-backend skill guides AI agents with real Gorig APIs, module boundaries, source checks, and framework rules.
4. Verify	Generated services are expected to pass go fmt, go vet, go build, tests, and route-level smoke checks.
5. Operate	Services can connect to operations tooling, runtime nodes, service registration, generated SDKs, and deployment workflows.

Ecosystem

Project	Role
gorig	Go backend framework: HTTP, routing, response helpers, domain/data access, cache, cron, messaging, SSE, auth, logging, and service lifecycle.
gorig-cli	Productive entry point: initialize projects, create modules, generate CRUD, produce docs, and install AI skills.
gorig-backend skill	AI delivery guide for Codex and Claude: source-aware implementation rules, testing policy, framework references, and scenario decomposition.
gorig-node	Hub-aware service node: register service handlers and expose direct or hub-routed calls.
gorig-hub	Service registry and SDK generator: receive node metadata, manage heartbeat, and publish generated Go SDKs.
gorig-om	Operations panel: service status, configuration, logs, API latency, error signatures, goroutine trends, and memory diagnostics.

Quick Start

Create a New Backend

Run without installing globally:

npx gorig-cli@latest init my-new-project --no-start

Or install the CLI globally:

npm install -g gorig-cli
gorig-cli init my-new-project --no-start

The generated project includes a runnable entry point, local/dev/prod configuration, and a dependency-light example module.

Run the Project

cd my-new-project
GORIG_SYS_MODE=local go run ./_cmd

Add a Module

npx gorig-cli@latest create user

This creates a flat feature module:

domain/user/
├── router.go
├── controller.go
├── service.go
├── dto.go
└── model/
    └── user.go

Generate Persistent CRUD

Choose the storage backend explicitly when you want database-backed CRUD:

npx gorig-cli@latest create order --crud --db mysql --db-name Main
npx gorig-cli@latest create order --crud --db mongo --db-name main

The CRUD generator creates service/model logic, optional HTTP routes, validation tests, module docs, API docs, and non-secret configuration skeletons.

Use with AI Agents

Install the bundled gorig-backend skill when you want Codex or Claude to work with Gorig projects using framework-aware rules instead of generic backend generation.

npx gorig-cli@latest skill install codex
npx gorig-cli@latest skill install all
npx gorig-cli@latest skill install codex project

Then ask for backend work in product language:

Use the gorig-backend skill to create a customer management backend with CRUD APIs, MySQL persistence, tests, and API docs.

Use the gorig-backend skill to add login, protected routes, token refresh, logout, and security tests to this Gorig service.

Use the gorig-backend skill to prepare this service for deployment with health checks, structured logs, release layout, and rollback steps.

Framework Installation

If you only need the Go framework dependency:

go get github.com/jom-io/gorig@latest

For most new projects, start with gorig-cli instead of adding the package manually.

Quality Gates

Recommended checks before shipping:

go fmt ./...
go vet ./...
go build ./...
go test ./... -v

For generated persistent CRUD modules, run the default tests first, then run database integration tests after local MySQL or MongoDB configuration is ready.