Go Api Design logo

Go Api Design

Community
eduardo-sl
go-api-design

HTTP/REST API design for Go services: handlers, ServeMux routing, middleware, request validation, pagination, idempotency, health checks, error responses, and graceful shutdown. Use when designing APIs, writing HTTP handlers, implementing middleware, or structuring REST endpoints. Trigger examples: "design API", "REST endpoints", "HTTP handler", "middleware pattern", "graceful shutdown", "API versioning". Not for: gRPC services (go-grpc), spec-first OpenAPI work (go-openapi), general architecture (go-architecture-review), concurrency in handlers (go-concurrency-review).

Overview

Publishereduardo-sl
Repositorygo-agent-skills
Skill namego-api-design
Stars
86
Forks
11
Bundled files
1
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.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

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

Installation

Install the Go Api Design 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-api-design
  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 Api Design 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 Api Design 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 Api Design 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 API Design

APIs are contracts. Once published, they're promises. Design them as if you'll maintain them for a decade — because you probably will.

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

Detailed reference material, loaded on demand:

  • references/handlers-and-middleware.md — handler structs, JSON response helpers, request ID and recoverer middleware, decode-and-validate.

Read a reference file only when the summary below is not enough.

1. HTTP Handler Structure

  • Route with the standard http.ServeMux (Go 1.22+): "GET /users/{id}" patterns match the method, r.PathValue("id") reads the wildcard, and a wrong method gets a 405 with Allow. A router dependency earns its place for middleware groups and sub-routers, not for path parameters.
  • Handlers are methods on a struct that holds their dependencies (store, logger) — never package globals.
  • Handler methods return nothing: they write the response and handle the error themselves, mapping domain errors (errors.Is(err, ErrNotFound)) to status codes and logging only the unexpected ones.
  • Route through one JSON helper that sets Content-Type, writes the status once, and logs encode failures.

2. Middleware Pattern

Middleware is func(http.Handler) http.Handler. Order, outside → inside:

text
Recoverer → RequestID → Logger → Auth → RateLimit → Handler

Recover MUST be outermost. Auth before business logic. Logger captures timing.

3. Request Validation

  • Cap the body before decoding: r.Body = http.MaxBytesReader(w, r.Body, 1<<20).
  • Decode into a typed request struct, then validate it in the same step — a successful decode is not a valid value.
  • Reply 400 with a stable message; never echo the decoder's error text.

4. URL and Naming Conventions

text
GET    /api/v1/users          → list users
POST   /api/v1/users          → create user
GET    /api/v1/users/{id}     → get user
PUT    /api/v1/users/{id}     → replace user
PATCH  /api/v1/users/{id}     → partial update
DELETE /api/v1/users/{id}     → delete user

GET    /api/v1/users/{id}/orders → list user orders (nested resource)

Rules:

  • Plural nouns for resources: /users, not /user
  • Kebab-case for multi-word paths: /order-items
  • One JSON field convention per API, never mixed — camelCase ("createdAt") is the default here; match an existing API if one exists
  • Version in URL path: /api/v1/...
  • No verbs in URLs: /users/search?q=alice, NOT /searchUsers

5. Pagination

go
type PageRequest struct {
    Cursor string `json:"cursor"`
    Limit  int    `json:"limit"`
}

type PageResponse[T any] struct {
    Items      []T    `json:"items"`
    NextCursor string `json:"nextCursor,omitempty"`
    HasMore    bool   `json:"hasMore"`
}

Prefer cursor-based pagination over offset/limit for large datasets. Offset pagination breaks under concurrent writes.

6. Graceful Shutdown

go
func main() {
    if err := run(context.Background()); err != nil {
        slog.Error("server stopped", slog.Any("error", err))
        os.Exit(1)
    }
}

func run(ctx context.Context) error {
    ctx, stop := signal.NotifyContext(ctx, os.Interrupt, syscall.SIGTERM)
    defer stop()

    srv := &http.Server{
        Addr:              ":8080",
        Handler:           router,
        ReadHeaderTimeout: 5 * time.Second, // Slowloris guard
        ReadTimeout:       10 * time.Second,
        WriteTimeout:      15 * time.Second,
        IdleTimeout:       120 * time.Second,
    }

    errc := make(chan error, 1)
    go func() { errc <- srv.ListenAndServe() }()

    select {
    case err := <-errc:
        return fmt.Errorf("listen: %w", err) // e.g. port already in use
    case <-ctx.Done():
    }

    // ctx is already cancelled; shutdown needs its own deadline.
    shutdownCtx, cancel := context.WithTimeout(context.WithoutCancel(ctx), 30*time.Second)
    defer cancel()
    if err := srv.Shutdown(shutdownCtx); err != nil {
        return fmt.Errorf("shutdown: %w", err)
    }
    if err := <-errc; !errors.Is(err, http.ErrServerClosed) {
        return err
    }
    return nil
}
  • ReadHeaderTimeout is the Slowloris defence: without it a client can hold a connection open by trickling header bytes. gosec G112 flags a server that lacks it.
  • Compare http.ErrServerClosed with errors.Is, never ==.
  • Exit only from main. A log.Fatal inside the serving goroutine skips every deferred cleanup, including the shutdown itself.

7. Health Check Endpoints

go
// Liveness: is the process alive? GET /healthz → 200, no dependency checks
// Readiness: should it receive traffic? GET /readyz → 200 or 503
func (h *HealthHandler) handleReady(w http.ResponseWriter, r *http.Request) {
    if h.draining.Load() { // atomic.Bool, set when shutdown starts
        h.respondError(w, http.StatusServiceUnavailable, "shutting down")
        return
    }
    if err := h.db.PingContext(r.Context()); err != nil {
        h.respondError(w, http.StatusServiceUnavailable, "database unavailable")
        return
    }
    h.respondJSON(w, http.StatusOK, map[string]string{"status": "ready"})
}

On SIGTERM behind a load balancer: set draining, wait one readiness period so the balancer stops routing, then call srv.Shutdown. Shutting down first rejects requests the balancer is still sending. Keep liveness free of dependency checks — a database outage should not make the orchestrator restart every replica.

8. Error Response Format

Consistent error responses across the entire API:

json
{
    "error": {
        "code": "VALIDATION_ERROR",
        "message": "invalid request parameters",
        "details": [
            {"field": "email", "message": "must be a valid email"}
        ]
    }
}

Map internal errors to HTTP status codes at the handler boundary. Internal errors should NEVER leak to clients. For a standard shape, use RFC 9457 problem details (application/problem+json) — the go-openapi skill has the type.

9. Idempotency and Browser Clients

  • POST endpoints that create or charge accept an Idempotency-Key header: store key → response for a bounded time, replay the stored response for a repeated key, and reject a concurrent duplicate with 409. Clients and proxies retry; without this, a retry is a double charge.
  • APIs called from browsers with cookies need CSRF protection — http.CrossOriginProtection (Go 1.25+); the go-security-audit skill owns the details.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Go Api Design AI skill do?

HTTP/REST API design for Go services: handlers, ServeMux routing, middleware, request validation, pagination, idempotency, health checks, error responses, and graceful shutdown. Use when designing APIs, writing HTTP handlers, implementing middleware, or structuring REST endpoints. Trigger examples: "design API", "REST endpoints", "HTTP handler", "middleware pattern", "graceful shutdown", "API versioning". Not for: gRPC services (go-grpc), spec-first OpenAPI work (go-openapi), general architecture (go-architecture-review), concurrency in handlers (go-concurrency-review).

Why use Go Api Design on TypingMind?

Because you install it once and use it with any model. Go Api Design 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 Api Design 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-api-design. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Go Api Design?

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 Api Design?

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

Is the Go Api Design 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 👇