Go Openapi logo

Go Openapi

Community
eduardo-sl
go-openapi

Spec-first REST development in Go with OpenAPI: generating server interfaces and clients with oapi-codegen, request validation middleware, RFC 9457 error bodies, contract testing, and detecting breaking API changes. Use when the API has or should have an OpenAPI document, when wiring code generation into the build, or when a hand-written handler has drifted from its published contract. Trigger examples: "OpenAPI", "swagger spec", "oapi-codegen", "generate a client from the spec", "validate requests against the schema", "breaking API change". Not for: handler structure and shutdown (go-api-design), gRPC and protobuf (go-grpc), GraphQL schemas and resolvers (go-graphql).

Overview

Publishereduardo-sl
Repositorygo-agent-skills
Skill namego-openapi
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 Openapi 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-openapi
  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 Openapi 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 Openapi 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 Openapi 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 OpenAPI

The spec is the source of truth. Types, routes, and clients are generated from it — never hand-written alongside it, because two hand-maintained copies of a contract diverge within one sprint.

Operating Modes

  • Adopt — the project has hand-written handlers and no spec, or a spec nobody generates from. Introduce generation without a rewrite.
  • Extend — the pipeline exists. Change the spec, regenerate, implement.
  • Review — check that handlers, spec, and published client agree, and that the change is not silently breaking.

1. Choose the Generator Once

ToolUse it when
oapi-codegenDefault. Types + server interface + client, works with net/http, chi, echo, gin
ogenYou want a fully generated, strictly validating server and can accept its opinions
swaggo/swagOnly for a legacy code-first project you are not converting — annotations generate the spec, so the spec cannot be reviewed before the code exists

Do not mix. A project with both annotations and a checked-in spec has two sources of truth again.

2. Wire Generation Into the Build

Pin the generator as a module tool (Go 1.24+), so every developer and CI run uses the same version:

bash
go get -tool github.com/oapi-codegen/oapi-codegen/v2/cmd/oapi-codegen
yaml
# oapi-codegen.yaml
package: api
output: internal/api/openapi.gen.go
generate:
  models: true
  std-http-server: true   # Go 1.22+ ServeMux; use chi-server / echo-server if that is the router
  strict-server: true     # typed request/response structs instead of raw http.ResponseWriter
  embedded-spec: true     # lets the service serve its own spec
output-options:
  skip-prune: false
go
//go:generate go tool oapi-codegen -config oapi-codegen.yaml ../../api/openapi.yaml

Commit generated files. Reviewers need to see the diff of a contract change, and a build must not depend on a generator being installed.

CI must fail when they are stale:

bash
go generate ./... && git diff --exit-code

3. Implement the Generated Interface

Strict mode gives typed requests and responses, so the compiler enforces the contract.

go
// Generated: type StrictServerInterface interface { GetUser(ctx, GetUserRequestObject) (GetUserResponseObject, error) }

type Server struct{ users UserStore }

var _ api.StrictServerInterface = (*Server)(nil) // compile-time compliance

func (s *Server) GetUser(ctx context.Context, req api.GetUserRequestObject) (api.GetUserResponseObject, error) {
    u, err := s.users.Find(ctx, req.Id)
    if errors.Is(err, ErrNotFound) {
        return api.GetUser404JSONResponse{Title: "user not found", Status: 404}, nil
    }
    if err != nil {
        return nil, fmt.Errorf("find user %s: %w", req.Id, err) // 500 via the error handler
    }
    return api.GetUser200JSONResponse{Id: u.ID, Email: u.Email}, nil
}

Rules:

  • Assert var _ api.StrictServerInterface = (*Server)(nil). Adding an endpoint to the spec then becomes a compile error, not a 404 in staging.
  • Return a typed response for every documented status. Return a Go error only for the undocumented failure path.
  • Never edit *.gen.go. Every change starts in the YAML.

4. Validate Requests Against the Spec

Generated types check shape, not constraints. minLength, pattern, enum, and required on query parameters are enforced only if you add the validation middleware.

go
spec, err := api.GetSwagger()
if err != nil {
    return fmt.Errorf("load spec: %w", err)
}
spec.Servers = nil // otherwise the server URL must match exactly

mux := http.NewServeMux()
handler := nethttpmiddleware.OapiRequestValidator(spec)(mux)

This rejects malformed input at the boundary with a 400 before any handler runs. Keep domain validation in the domain — the middleware enforces the contract, not the business rules.

5. Errors: RFC 9457 Problem Details

Define one error schema and reference it from every failure response.

yaml
components:
  schemas:
    Problem:
      type: object
      required: [type, title, status]
      properties:
        type:   { type: string, format: uri, default: "about:blank" }
        title:  { type: string }
        status: { type: integer }
        detail: { type: string }
        instance: { type: string }

Serve it as application/problem+json. Never return a bare string, and never put an internal error message in detail — log the wrapped error, return a stable, safe title.

6. Versioning and Breaking Changes

Detect breaking changes mechanically; reviewers miss them.

bash
go install github.com/oasdiff/oasdiff@latest
oasdiff breaking api/openapi.yaml.base api/openapi.yaml --fail-on ERR

Run it in CI against the spec on the main branch. Breaking, in practice: removing an endpoint or field, narrowing a type, adding a required request field or a required response field the client must understand, changing a status code, removing an enum value from a response.

Additive changes are safe. Version the path (/v2/...) only when a break is unavoidable, and keep the previous version serving until clients have moved.

7. Contract Testing

The spec is only a contract if something fails when the implementation disagrees.

go
func TestGetUser_MatchesSpec(t *testing.T) {
    spec, err := api.GetSwagger()
    require.NoError(t, err)
    spec.Servers = nil

    srv := httptest.NewServer(newTestHandler(t, spec))
    t.Cleanup(srv.Close)

    // Generated client — if the spec changed, this stops compiling
    c, err := api.NewClientWithResponses(srv.URL)
    require.NoError(t, err)

    resp, err := c.GetUserWithResponse(t.Context(), "u-1")
    require.NoError(t, err)
    require.Equal(t, http.StatusOK, resp.StatusCode())
    require.Equal(t, "u-1", resp.JSON200.Id)
}

Use the generated client in tests, not a hand-rolled http.NewRequest. It turns contract drift into a compile failure.

8. Keep the Spec Reviewable

bash
vacuum lint -d api/openapi.yaml     # or: spectral lint, redocly lint
  • One file per API, under api/, checked in, reviewed like code.
  • Every operation has an operationId — it becomes the Go method name.
  • Every schema has a description; it becomes the godoc on the generated type.
  • Split large specs with $ref to components/, not by generating fragments.

Verification Checklist

  1. Exactly one source of truth: a checked-in spec, no annotation generator alongside it
  2. The generator is pinned via the go.mod tool directive
  3. Generated files are committed and CI fails on go generate + git diff --exit-code
  4. var _ api.StrictServerInterface = (*Server)(nil) present
  5. No hand edits in *.gen.go
  6. Request validation middleware installed and covered by a 400 test
  7. Errors use a single Problem schema, served as application/problem+json
  8. oasdiff breaking runs in CI against the base spec
  9. At least one test drives the generated client against the real handler
  10. The spec lints clean and every operation has an operationId

Frequently asked questions

What does the Go Openapi AI skill do?

Spec-first REST development in Go with OpenAPI: generating server interfaces and clients with oapi-codegen, request validation middleware, RFC 9457 error bodies, contract testing, and detecting breaking API changes. Use when the API has or should have an OpenAPI document, when wiring code generation into the build, or when a hand-written handler has drifted from its published contract. Trigger examples: "OpenAPI", "swagger spec", "oapi-codegen", "generate a client from the spec", "validate requests against the schema", "breaking API change". Not for: handler structure and shutdown (go-api-d...

Why use Go Openapi on TypingMind?

Because you install it once and use it with any model. Go Openapi 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 Openapi 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-openapi. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Go Openapi?

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 Openapi?

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

Is the Go Openapi 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 👇