golang-grpc
npx skills add https://github.com/bobmatnyc/claude-mpm-skills --skill golang-grpc
Agent 安装分布
Skill 文档
Go gRPC (Production)
Overview
gRPC provides strongly-typed RPC APIs backed by Protocol Buffers, with first-class streaming support and excellent performance for service-to-service communication. This skill focuses on production defaults: versioned protos, deadlines, error codes, interceptors, health checks, TLS, and testability.
Quick Start
1) Define a versioned protobuf API
â Correct: versioned package
// proto/users/v1/users.proto
syntax = "proto3";
package users.v1;
option go_package = "example.com/myapp/gen/users/v1;usersv1";
service UsersService {
rpc GetUser(GetUserRequest) returns (GetUserResponse);
rpc ListUsers(ListUsersRequest) returns (stream User);
}
message GetUserRequest { string id = 1; }
message GetUserResponse { User user = 1; }
message ListUsersRequest { int32 page_size = 1; string page_token = 2; }
message User {
string id = 1;
string email = 2;
string display_name = 3;
}
â Wrong: unversioned package (hard to evolve)
package users;
2) Generate Go code
Install generators:
go install google.golang.org/protobuf/cmd/protoc-gen-go@latest
go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@latest
Generate:
protoc -I proto \
--go_out=./gen --go_opt=paths=source_relative \
--go-grpc_out=./gen --go-grpc_opt=paths=source_relative \
proto/users/v1/users.proto
3) Implement server with deadlines and status codes
â Correct: validate + map errors to gRPC codes
package usersvc
import (
"context"
"google.golang.org/grpc/codes"
"google.golang.org/grpc/status"
usersv1 "example.com/myapp/gen/users/v1"
)
type Service struct {
usersv1.UnimplementedUsersServiceServer
Repo Repo
}
type Repo interface {
GetUser(ctx context.Context, id string) (User, error)
}
type User struct {
ID, Email, DisplayName string
}
func (s *Service) GetUser(ctx context.Context, req *usersv1.GetUserRequest) (*usersv1.GetUserResponse, error) {
if req.GetId() == "" {
return nil, status.Error(codes.InvalidArgument, "id is required")
}
u, err := s.Repo.GetUser(ctx, req.GetId())
if err != nil {
if err == ErrNotFound {
return nil, status.Error(codes.NotFound, "user not found")
}
return nil, status.Error(codes.Internal, "internal error")
}
return &usersv1.GetUserResponse{
User: &usersv1.User{
Id: u.ID,
Email: u.Email,
DisplayName: u.DisplayName,
},
}, nil
}
â Wrong: return raw errors (clients lose code semantics)
return nil, errors.New("user not found")
Core Concepts
Deadlines and cancellation
Make every call bounded; enforce server-side timeouts for expensive handlers.
â Correct: require deadline
if _, ok := ctx.Deadline(); !ok {
return nil, status.Error(codes.InvalidArgument, "deadline required")
}
Metadata
Use metadata for auth/session correlation, not for primary request data.
â Correct: read auth token from metadata
md, _ := metadata.FromIncomingContext(ctx)
auth := ""
if vals := md.Get("authorization"); len(vals) > 0 {
auth = vals[0]
}
Interceptors (Middleware)
Use interceptors for cross-cutting concerns: auth, logging, metrics, tracing, request IDs.
â Correct: unary interceptor with request ID
func unaryRequestID() grpc.UnaryServerInterceptor {
return func(ctx context.Context, req any, info *grpc.UnaryServerInfo, handler grpc.UnaryHandler) (any, error) {
id := uuid.NewString()
ctx = context.WithValue(ctx, requestIDKey{}, id)
resp, err := handler(ctx, req)
return resp, err
}
}
Streaming patterns
Server streaming (paginate or stream results)
â Correct: stop on ctx.Done()
func (s *Service) ListUsers(req *usersv1.ListUsersRequest, stream usersv1.UsersService_ListUsersServer) error {
users, err := s.Repo.ListUsers(stream.Context(), int(req.GetPageSize()))
if err != nil {
return status.Error(codes.Internal, "internal error")
}
for _, u := range users {
select {
case <-stream.Context().Done():
return stream.Context().Err()
default:
}
if err := stream.Send(&usersv1.User{
Id: u.ID,
Email: u.Email,
DisplayName: u.DisplayName,
}); err != nil {
return err
}
}
return nil
}
Unary vs streaming decision
- Use unary for single request/response and simple retries.
- Use server streaming for large result sets or continuous updates.
- Use client streaming for bulk uploads with one final response.
- Use bidirectional streaming for interactive protocols.
Production Hardening
Health checks and reflection
Add health service; enable reflection only in non-production environments.
â Correct: health + conditional reflection
hs := health.NewServer()
grpc_health_v1.RegisterHealthServer(s, hs)
if env != "production" {
reflection.Register(s)
}
Graceful shutdown
Prefer GracefulStop with a deadline.
â Correct: graceful stop
stopped := make(chan struct{})
go func() {
grpcServer.GracefulStop()
close(stopped)
}()
select {
case <-stopped:
case <-time.After(10 * time.Second):
grpcServer.Stop()
}
TLS
Use TLS (or mTLS) in production; avoid insecure credentials outside local dev.
â Correct: server TLS
creds, err := credentials.NewServerTLSFromFile("server.crt", "server.key")
if err != nil { return err }
grpcServer := grpc.NewServer(grpc.Creds(creds))
Testing (bufconn)
Test gRPC handlers without opening real sockets using bufconn.
â Correct: in-memory gRPC test server
const bufSize = 1024 * 1024
lis := bufconn.Listen(bufSize)
srv := grpc.NewServer()
usersv1.RegisterUsersServiceServer(srv, &Service{Repo: repo})
go func() { _ = srv.Serve(lis) }()
ctx := context.Background()
conn, err := grpc.DialContext(
ctx,
"bufnet",
grpc.WithContextDialer(func(context.Context, string) (net.Conn, error) { return lis.Dial() }),
grpc.WithTransportCredentials(insecure.NewCredentials()),
)
if err != nil { t.Fatal(err) }
defer conn.Close()
client := usersv1.NewUsersServiceClient(conn)
resp, err := client.GetUser(ctx, &usersv1.GetUserRequest{Id: "1"})
_ = resp
_ = err
Anti-Patterns
-
Ignore deadlines: unbounded handlers cause tail latency and resource exhaustion.
-
Return string errors: map domain errors to
codes.*withstatus.Errororstatus.Errorf. -
Stream without backpressure: stop on
ctx.Done()and handleSenderrors. -
Expose reflection in production: treat reflection as a discovery surface.
Troubleshooting
Symptom: clients see UNKNOWN errors
Actions:
- Return
status.Error(codes.X, "...")instead of raw errors. - Wrap domain errors into typed errors, then map to gRPC codes.
Symptom: slow/hanging requests
Actions:
- Require deadlines and propagate
ctxto downstream calls. - Add server-side timeouts and bounded concurrency in repositories.
Symptom: flaky streaming
Actions:
- Stop streaming on
ctx.Done()and handlestream.Senderrors. - Avoid buffering entire result sets before sending.
Resources
- gRPC Go: https://github.com/grpc/grpc-go
- Protobuf Go: https://pkg.go.dev/google.golang.org/protobuf
- gRPC error codes: https://grpc.io/docs/guides/error/