Go Project Layout logo

Go Project Layout

Community
eduardo-sl
go-project-layout

Scaffold new Go projects and services: directory structure, cmd/ and internal/ conventions, when to use a flat layout, module naming, and main package wiring. Use when: "new Go project", "scaffold a service", "create project structure", "start a Go module", "how do I organize a new service", "set up folder structure". Not for: reviewing an existing architecture (go-architecture-review), DI wiring (go-dependency-injection), CI setup (go-ci).

Overview

Publishereduardo-sl
Repositorygo-agent-skills
Skill namego-project-layout
Stars
86
Forks
11
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by eduardo-sl on GitHub. Read the source before you install it.

Installation

Install the Go Project Layout AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/eduardo-sl/go-agent-skills/tree/main/skills/(architecture)/go-project-layout
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Use it in TypingMind

Enable Go Project Layout in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Go Project Layout on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Go Project Layout is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Go Project Layout

Structure follows size. The biggest layout mistake in Go is copying a microservice skeleton for a 500-line tool — or growing a 50-package service inside a flat directory. Match the layout to the project.

Community default. A project skill or agent-config rule that explicitly supersedes go-project-layout takes precedence over this one.

1. Pick the Layout by Project Size

ProjectLayout
Small tool, single binary, <5 filesFlat: everything in package main at the root
Library for others to importRoot package named after the module, internal/ for helpers
Service with one binarycmd/<name>/main.go + internal/ packages
Multiple binaries sharing codecmd/<name1>/, cmd/<name2>/ + internal/

Never start with empty pkg/, api/, docs/, build/ directories "for later". Add structure when the code demands it, not before.

2. Module Naming

bash
# ✅ Good — repository path, lowercase
go mod init github.com/acme/payment-service

# ❌ Bad — not fetchable, uppercase, or vanity without DNS
go mod init PaymentService
go mod init payment_service

The last path element should match what users will see: for a library, it becomes the default import name.

3. Service Layout (the default for APIs and workers)

text
payment-service/
├── cmd/
│   └── payment-api/
│       └── main.go         # flag/env parsing, wiring, Run() — nothing else
├── internal/
│   ├── domain/             # core types, business rules; zero external deps
│   ├── service/            # use cases orchestrating domain + stores
│   ├── store/              # data access implementations (postgres/, redis/)
│   ├── handler/            # HTTP/gRPC adapters
│   └── config/             # config loading and validation
├── migrations/             # if the service owns a database
├── go.mod
├── Makefile
└── README.md

Rules:

  • internal/ by default — the compiler enforces that nobody outside the module imports it. Promote to a public package only on demand.
  • pkg/ only when external consumers exist AND the module also has private code. When in doubt, don't create it.
  • Dependencies point inward: handler → service → domain ← store. domain imports neither store nor handler.

4. Thin main, Runnable Run

Keep main.go to wiring plus a delegating call, so the app is testable:

go
func main() {
    if err := run(context.Background(), os.Args[1:], os.Getenv); err != nil {
        fmt.Fprintln(os.Stderr, err)
        os.Exit(1)
    }
}

func run(ctx context.Context, args []string, getenv func(string) string) error {
    cfg, err := config.Load(getenv)
    if err != nil {
        return fmt.Errorf("load config: %w", err)
    }

    db, err := store.Open(ctx, cfg.DatabaseURL)
    if err != nil {
        return fmt.Errorf("open db: %w", err)
    }
    defer db.Close()

    svc := service.New(store.NewUserRepo(db))
    srv := handler.NewServer(cfg.Addr, svc)
    return srv.ListenAndServe(ctx)
}
  • os.Exit appears exactly once, in main.
  • run takes its dependencies (args, getenv) so tests can call it.
  • No init() functions for wiring — explicit construction order only.

5. Library Layout

text
retry/
├── retry.go            # package retry — the API, in the root
├── retry_test.go
├── backoff.go          # same package, split by topic
├── internal/
│   └── clock/          # implementation details users must not import
├── examples_test.go    # Example* functions shown in godoc
└── go.mod
  • The root directory IS the package. No src/, no lib/.
  • One package per concept. Resist util, common, helpers — name packages after what they provide (retry, clock, httpsign).

6. Naming Rules for Directories and Packages

  • Package name == directory name, short, lowercase, no underscores: store/postgres, not store/postgres_impl.
  • Don't stutter: payment.Service, not payment.PaymentService.
  • Binary names in cmd/ are user-facing: cmd/payment-api, hyphenated is fine (directory only holds package main).

7. Files That Belong at the Root

  • go.mod, go.sum, README.md, LICENSE, Makefile, .golangci.yml, Dockerfile (single-binary projects).
  • Do NOT create: src/ (un-idiomatic), vendor/ (unless the team explicitly vendors), one-file packages like types/ or models/ that become dumping grounds.

8. Configuration

  • One config struct, loaded from the environment in one place (internal/config), validated at startup — fail fast with every missing or invalid value listed, not the first one.
  • Flags for CLIs; environment variables for services (12-factor); defaults in code. No config files baked into images, no environment-specific branches (if env == "prod") in business code.
  • Secrets come from the environment or a secret manager, never from flags (ps shows argv) or the repository.

9. One Module or Several

Default to one module per repository. Split into modules only when parts are versioned and released independently (a public client library next to a service). Develop across modules locally with go.work; don't commit replace => ../path directives.

Scaffolding Procedure

  1. Ask/decide: tool, library, or service? How many binaries?
  2. go mod init <repo-path>, then set the go directive deliberately — since Go 1.26, go mod init writes the previous release (go 1.(N-1).0), not the toolchain's.
  3. Create only the directories the first feature needs.
  4. Write main.go with the thin-main pattern above, and a config struct.
  5. Pin developer tools as module tools (go get -tool ..., Go 1.24+) and add Makefile targets: build, test, lint.
  6. Verify: go build ./... and go vet ./... pass on the skeleton.

Verification Checklist

  1. Layout matches project size — no empty scaffolding directories
  2. Module path is the fetchable repository path
  3. All non-public packages live under internal/
  4. main.go is thin: parse, wire, call run, exit
  5. os.Exit only in main; no wiring in init()
  6. Dependencies flow inward; domain has zero infrastructure imports
  7. No util/common/helpers/models grab-bag packages
  8. Package names match directories, lowercase, no stutter
  9. go build ./... passes on the fresh skeleton
  10. Config loaded and validated in one place at startup; no secrets in flags
  11. One module unless parts are released independently; no committed local replace

Frequently asked questions

What does the Go Project Layout AI skill do?

Scaffold new Go projects and services: directory structure, cmd/ and internal/ conventions, when to use a flat layout, module naming, and main package wiring. Use when: "new Go project", "scaffold a service", "create project structure", "start a Go module", "how do I organize a new service", "set up folder structure". Not for: reviewing an existing architecture (go-architecture-review), DI wiring (go-dependency-injection), CI setup (go-ci).

Why use Go Project Layout on TypingMind?

Because you install it once and use it with any model. Go Project Layout is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Go Project Layout in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/eduardo-sl/go-agent-skills/tree/main/skills/(architecture)/go-project-layout. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Go Project Layout?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Go Project Layout?

As many as you like. As long as a model supports skills, you can use Go Project Layout with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Go Project Layout AI skill free?

Yes. It is published on GitHub by eduardo-sl under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇