CAT
/MCP
SkillsMCPMarketplacesDigestToolsAdvertise

This week in Claude

Every Monday: Claude Code, Agent SDK, MCP, and the Anthropic platform moves worth your time.

Skills by Category
Frontend DevelopmentBackend & APIsTesting & QASecurityDevOps & CI/CDGit & Pull RequestsDocumentationCode Review & QualityAI & Agent BuildingSkill Development
MCP Servers by Category
Sales & MarketingWeb & Browser AutomationDatabasesAI & LLM ToolsCloud & InfrastructureCommunication & MessagingDeveloper ToolsDesign & CreativeDocuments & KnowledgeSearch & Web Crawling
Marketplaces by Category
AI Agents & OrchestrationLLM IntegrationDevelopment ToolsFrontend & UIBackend & APIsDatabasesTesting & Code QualityDevOps & CloudSecurity & ComplianceGit & Version Control

Cross AI Tools

Discover Claude Code plugins, extensions, and tools. Automatically updated directory of Anthropic Claude AI marketplaces with development tools, productivity plugins, and integrations.

Resources

  • Browse Skills
  • Browse MCP Servers
  • Browse Marketplaces
  • Plugins Reference

Community

  • About
  • Tools
  • Feedback
  • Privacy Policy
  • Advertise

Built for the Claude Code community with Claude Code by @mertduzgun

Independent project, not affiliated with Anthropic

Tiny Go MCP Server

kioie/tiny-go-mcp-server
1STDIOregistry active
Summary

A minimal reference implementation for testing the tinymcp Go toolkit, not meant for production workloads. Exposes basic tools, resources, and prompts over stdio to verify that the lightweight MCP wrapper correctly generates JSON schemas from struct tags, handles registration, and manages transport boilerplate on top of the official modelcontextprotocol/go-sdk. Useful if you're building your own Go MCP server with tinymcp and want a working example to compare against, or if you need a throwaway server to test client integration without side effects. Ships as a container image at ghcr.io/kioie/tiny-go-mcp for quick iteration.

CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
Keep your Mac awake
Keep your Mac awake
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →
Context.devContext.dev
Context.dev
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Make money from your Skills
Make money from your Skills
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
AppSignal
AppSignal
Monitor with ease. Code with confidence.
Start Free Trial →
CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
Keep your Mac awake
Keep your Mac awake
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →
Context.devContext.dev
Context.dev
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Make money from your Skills
Make money from your Skills
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
AppSignal
AppSignal
Monitor with ease. Code with confidence.
Start Free Trial →

Tiny Go MCP Server — minimal Go MCP toolkit

Tiny Go MCP Server

CI Go Reference Go Report Card tiny-go-mcp-server MCP server

A lightweight Model Context Protocol (MCP) toolkit for Go. Build spec-compliant MCP servers (stdio, streamable HTTP, legacy SSE) with tools, resources, and prompts — minimal boilerplate and automatic JSON Schema from Go structs.

Built on the official modelcontextprotocol/go-sdk.

Requirements: Go 1.26+ (download).


Why Tiny?

Tiny Go MCP ServerFull frameworks
GoalThin helper on official go-sdk + tiny static binaryFull MCP feature surface
DepsOfficial go-sdk onlyVaries
Binary~5MB stripped, no runtime on hostOften larger stacks
SchemasInferred from struct tagsManual or builder APIs

Use this project as a library (tinymcp package) or as a starting template (cmd/tiny-go-mcp).

When to use what

tinymcp (this repo)mcp-gogo-sdk alone
Best forThin helper on go-sdk, tiny binaryRich helpers, large ecosystemFull control, no extra layer
SchemaStruct tags → auto JSON SchemaBuilder APIs / helpersAddTool + generics yourself
Transportstdio (Start()), streamable HTTP (StartHTTP), legacy SSE (StartSSE)stdio, SSE, HTTP, …All transports
Depsgo-sdk onlyStandalone modulego-sdk only

Choose tinymcp when you want the official protocol implementation with minimal boilerplate and a small static server binary.

Philosophy

tinymcp is a thin helper on the official go-sdk — not a replacement for it.

We reduce setup and transport boilerplate (server creation, registration error handling, stdio/HTTP/SSE, TextResult, deploy examples). The protocol implementation, generics, and schema inference still come from modelcontextprotocol/go-sdk.

That means handler code uses both imports — and that is intentional:

import (
    "github.com/kioie/tiny-go-mcp-server/tinymcp"
    "github.com/modelcontextprotocol/go-sdk/mcp" // handler types, prompts, resources, advanced APIs
)
Use tinymcp forUse go-sdk (mcp) for
NewServer, RegisterTool, transportsHandler signatures (CallToolRequest, GetPromptRequest, …)
Safe registration (errors, not panics)Tool annotations, elicitation, custom protocol features
TextResult, HTTP middleware helpersAnything via server.RawServer()

We are not aiming for a non-leaky facade that hides the SDK. If you need full control, call RawServer() or use go-sdk directly — same underlying server, no lock-in.

tinymcp vs raw go-sdk

Same protocol implementation — tinymcp removes repetitive setup. Handler code still imports mcp for request types in both cases.

go-sdk alone (minimal stdio server):

server := mcp.NewServer(&mcp.Implementation{Name: "my-mcp", Version: "1.0.0"}, nil)
mcp.AddTool(server, &mcp.Tool{
    Name:        "greet",
    Description: "Greet someone by name",
}, greet)
if err := server.Run(context.Background(), &mcp.StdioTransport{}); err != nil {
    log.Fatal(err)
}

tinymcp (same tool, less boilerplate):

s := tinymcp.NewServer("my-mcp", "1.0.0")
if err := tinymcp.RegisterTool(s, "greet", "Greet someone by name", greet); err != nil {
    log.Fatal(err)
}
log.Fatal(s.Start())
tinymcp addsStill on go-sdk (mcp)
NewServer(name, ver)Handler signatures (CallToolRequest, prompts, resources)
RegisterTool + struct-tag JSON SchemaTool annotations via RegisterToolDef + mcp.Tool
Safe registration errors (no panics)Advanced session / event-store APIs
Start() / StartHTTP() / HTTP middlewareFull control via RawServer()

Use go-sdk alone when you want zero wrapper. Use tinymcp when you want less setup while staying on the official implementation.

Transport

MethodAPITypical clients
stdio (default)Start()Cursor, Claude Desktop, Windsurf (local subprocess)
Streamable HTTPStartHTTP(addr, opts) or StreamableHTTPHandlerRemote MCP clients, gateways, browser tools
Legacy SSEStartSSE(addr, opts) or SSEHandlerOlder clients on MCP 2024-11-05 SSE transport

Start() runs stdio (stdin/stdout) — what most local AI clients expect.

For HTTP/SSE, tinymcp wraps the official go-sdk handlers with minimal options:

// Streamable HTTP on loopback (stateless demo — no GET/SSE or server→client RPC)
log.Fatal(server.StartHTTP("127.0.0.1:8080", &tinymcp.HTTPOptions{Stateless: true}))

// Or mount on your own mux (auth, TLS, path prefix)
handler, _ := tinymcp.StreamableHTTPHandler(server, nil)
http.Handle("/mcp", handler)

Stateless mode (Stateless: true) is the default in examples: one POST JSON-RPC per request, no long-lived SSE GET stream, and no server-initiated messages. Omit it or use session options when you need full streamable HTTP sessions — see docs/HTTP.md.

See docs/HTTP.md and examples/http. To host for Smithery URL listing (no Docker for end users), use examples/http-deploy. For advanced session routing or event stores, use server.RawServer() with the go-sdk directly.


Quick start (library)

Step-by-step guide: docs/QUICKSTART.md. AI codegen: SYSTEM_PROMPT.md.

go get github.com/kioie/tiny-go-mcp-server/tinymcp@latest
package main

import (
	"context"
	"fmt"
	"log"

	"github.com/kioie/tiny-go-mcp-server/tinymcp"
	"github.com/modelcontextprotocol/go-sdk/mcp"
)

type greetArgs struct {
	Name string `json:"name" jsonschema:"Person to greet"`
}

func main() {
	s := tinymcp.NewServer("my-mcp", "1.0.0")
	if err := tinymcp.RegisterTool(s, "greet", "Greet someone by name", greet); err != nil {
		log.Fatal(err)
	}
	log.Fatal(s.Start())
}

func greet(_ context.Context, _ *mcp.CallToolRequest, args greetArgs) (*mcp.CallToolResult, any, error) {
	return tinymcp.TextResult(fmt.Sprintf("Hello, %s!", args.Name)), nil, nil
}

See examples/minimal for a runnable copy-paste example.

Scaffold a new server

Requires tagged module template/ (v1.1.1+) for stdio, or template-http/ for streamable HTTP:

go install golang.org/x/tools/cmd/gonew@latest
gonew github.com/kioie/tiny-go-mcp-server/template@latest example.com/my-mcp my-mcp
cd my-mcp && go run .

# HTTP deploy (Smithery / Fly / Render):
gonew github.com/kioie/tiny-go-mcp-server/template-http@latest example.com/my-mcp-http my-mcp-http
cd my-mcp-http && go run .

Or copy examples/minimal, template/, or template-http/ directly.


Install the example server

Both install paths produce a binary named tiny-go-mcp:

# Installs to $(go env GOPATH)/bin/tiny-go-mcp
go install github.com/kioie/tiny-go-mcp-server/cmd/tiny-go-mcp@latest

Or build from source (binary in the repo root):

git clone https://github.com/kioie/tiny-go-mcp-server.git
cd tiny-go-mcp-server
make release   # → ./tiny-go-mcp
MethodBinary nameTypical path
go install …/cmd/tiny-go-mcptiny-go-mcp$(go env GOPATH)/bin/tiny-go-mcp
make build / make releasetiny-go-mcp./tiny-go-mcp in the repo
make installtiny-go-mcp$(go env GOPATH)/bin/tiny-go-mcp

Example tools (reference server)

These tools exist for MCP integration demos, not production logic. Agents should compute math and write greetings in-chat unless they are explicitly testing tool calls.

ToolWhen to useWhen not to / alternativeArguments
addTest that the client can call an addition toolReal arithmetic → compute locally or use a calculator MCPa, b
subtractTest subtraction wiring (use instead of add for subtraction tests)Real arithmetic → compute locallya, b
greetTest a text-returning tool (use instead of add/subtract for messaging demos)User-facing hello → reply in the conversationname (required), greeting (optional)

Connect AI clients

MCP servers communicate over stdio. Point your client at the compiled binary path.

Template config: examples/mcp-client-config.json (copy and set the absolute path to tiny-go-mcp).

Logging: The protocol uses stdin/stdout. Server logs (if any) go to stderr only. Set TINY_GO_MCP_VERBOSE=1 on the server process to enable startup log lines.

Cursor

Settings → Features → MCP → Add server:

  • Name: tiny-go-mcp
  • Type: stdio
  • Command: /absolute/path/to/tiny-go-mcp

Or add to .cursor/mcp.json in your project:

{
  "mcpServers": {
    "tiny-go-mcp": {
      "command": "/absolute/path/to/tiny-go-mcp"
    }
  }
}

Claude Desktop

~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "tiny-go-mcp": {
      "command": "/absolute/path/to/tiny-go-mcp"
    }
  }
}

Windsurf / Zed / other stdio clients

Use the same shape: command = absolute path to tiny-go-mcp, transport = stdio. Refer to your client’s MCP docs for the config file location.

Tips for LLM-friendly tools

  • Use clear tool names (snake_case) and descriptions that say when to use, when not to, and which sibling tool applies — models pick tools from these and often have overlapping options.
  • Add jsonschema tags on struct fields so argument docs appear in the schema.
  • Return human-readable text via tinymcp.TextResult for predictable client display.
  • See AGENTS.md for conventions when extending this repo with AI assistants.

Resources and prompts

Register read-only context and reusable prompt templates alongside tools:

if err := tinymcp.RegisterTextResource(server, "file:///info", "info", "Server metadata", "text/plain", "…"); err != nil {
	log.Fatal(err)
}
if err := tinymcp.RegisterPrompt(server, "code_review", "Review code", []*mcp.PromptArgument{
	{Name: "code", Required: true},
}, func(_ context.Context, req *mcp.GetPromptRequest) (*mcp.GetPromptResult, error) {
	code := req.Params.Arguments["code"]
	if code == "" {
		return nil, tinymcp.RequiredPromptArgument("code")
	}
	return tinymcp.PromptResult("Review", tinymcp.UserPromptMessage("Review:\n"+code)), nil
}); err != nil {
	log.Fatal(err)
}

Runnable example: examples/resources. For dynamic URI templates use RegisterResourceTemplate.


Package API

server := tinymcp.NewServer("name", "version")
tinymcp.NewServer("name", "version", tinymcp.WithInstructions("…")) // optional SDK config
tinymcp.NewServerWithOptions("name", "version", &mcp.ServerOptions{…})
tinymcp.RegisterTool(server, name, description, handler)     // typed handler, auto schema
tinymcp.RegisterTextResource(server, uri, name, desc, mime, text)
tinymcp.RegisterPrompt(server, name, desc, args, handler)
server.Start()                                              // stdio transport
server.StartHTTP(":8080", &tinymcp.HTTPOptions{})         // streamable HTTP
server.StartSSE(":8080", nil)                             // legacy SSE
tinymcp.StreamableHTTPHandler(server, nil)                // mount on custom http.Server
tinymcp.TextResult("message")                               // tool text helper
tinymcp.TextResource(uri, mime, text)                       // resource read helper
tinymcp.PromptResult(desc, tinymcp.UserPromptMessage("…")) // prompt helper
server.RawServer()                                          // escape hatch to go-sdk

// v1.2+: panic-at-startup registration or errors.Is sentinels
tinymcp.MustRegisterTool(server, name, description, handler)
errors.Is(err, tinymcp.ErrNilServer)                        // ErrNilTool, ErrNilHandler, ErrRegistrationFailed
opts := (&tinymcp.HTTPOptions{Stateless: true}).WithMiddleware(requestLogger)
tinymcp.ListenAndServeHTTPContext(ctx, addr, handler)       // graceful shutdown; also StartHTTPContext / StartSSEContext

Documentation: pkg.go.dev/github.com/kioie/tiny-go-mcp-server/tinymcp. Upgrading from v1.1.x: docs/MIGRATION-v1.2.md.


Development

CommandDescription
make testRun tests with race detector
make lintgolangci-lint
make lint-toolsValidate MCP tool descriptions in reference servers
make coverageCoverage report
make buildDev binary ./tiny-go-mcp
make releaseStripped static binary
make installgo install → $(go env GOPATH)/bin/tiny-go-mcp

Smaller binaries (~1.8MB)

After make release, optionally pack with UPX:

upx --best --lzma tiny-go-mcp

Project structure

tinymcp/              # Library package
cmd/tiny-go-mcp/      # Reference MCP server
template/             # gonew stdio scaffold
template-http/        # gonew streamable HTTP deploy scaffold
examples/minimal/     # Minimal stdio example
examples/http/        # Streamable HTTP example
examples/http-deploy/ # Deployable HTTP + Smithery URL listing (server card, Render/Fly)
examples/resources/   # Resources + prompts example
examples/mcp-client-config.json  # Cursor/Claude-style template
scripts/lint-tools/   # MCP tool description linter (make lint-tools)
docs/                 # Guides — see below
server.json           # MCP Registry metadata (publish with mcp-publisher)
CHANGELOG.md          # Release history
SYSTEM_PROMPT.md      # Agent-facing API summary for codegen
.github/workflows/    # CI, lint, CodeQL, releases

Key docs in docs/:

DocPurpose
QUICKSTART.mdStep-by-step library setup
HTTP.mdstdio vs streamable HTTP vs legacy SSE
STABILITY.mdPublic API stability policy
MIGRATION-v1.2.mdUpgrade guide from v1.1.x
TLS.mdHTTPS via reverse proxy or Go
LOCALHOST-PROTECTION.mdDNS rebinding security advisory
DISCOVERY.mdRegistries and visibility
GLAMA.mdGlama hosting
SMITHERY.mdSmithery URL and MCPB listings

Releases

Tag a semver version (e.g. v1.2.0) to publish stable go get versions and trigger GitHub Releases with cross-platform binaries and multi-arch GHCR images. Release history: CHANGELOG.md. Public API stability: docs/STABILITY.md. Agent-facing API summary: SYSTEM_PROMPT.md. Upgrading from v1.1.x: docs/MIGRATION-v1.2.md.

git tag v1.2.0
git push origin v1.2.0

Discovery and registries

See docs/DISCOVERY.md for MCP Registry (server.json), awesome lists, and community directories. Listing copy and launch posts: docs/SUBMISSIONS.md. For Glama hosting with Docker, see docs/GLAMA.md.

Contributing

See CONTRIBUTING.md. For AI codegen outside this repo, see SYSTEM_PROMPT.md. CI runs tests, lint, and CodeQL; Dependabot keeps Go and Actions dependencies updated.

License

MIT — see LICENSE.

Featured
CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
Keep your Mac awake
Keep your Mac awake
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →
Context.devContext.dev
Context.dev
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Make money from your Skills
Make money from your Skills
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
AppSignal
AppSignal
Monitor with ease. Code with confidence.
Start Free Trial →
Categories
AI & LLM Tools
Registryactive
Packageghcr.io/kioie/tiny-go-mcp:1.1.2
TransportSTDIO
UpdatedMay 22, 2026
View on GitHub

Related AI & LLM Tools MCP Servers

View all →
SkillFM LLM Cost Optimizer

io.github.ericm1018/skillfm-llm-cost-optimizer-openai-anthropic-usage

LLM cost optimizer for OpenAI, Anthropic, token usage, BYOK, and SkillFM Beacon audits.
Llm Orchestration Agent

io.github.mikerawsonnz/llm-orchestration-agent

Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith
Authenticated Llm Agent

io.github.mikerawsonnz/authenticated-llm-agent

JWT-gated LLM gateway: authenticate (bcrypt/JWT), then run a LangChain-on-Vertex Gemini completion.
Copilot Memory MCP

labforgedev/copilot-memory-mcp

Persistent semantic memory for AI agents using local ChromaDB vector search. No cloud required.
1
Agent Prompt Injection Firewall Mcp

csoai-org/agent-prompt-injection-firewall-mcp

The WAF for agents. Pattern-based + heuristic firewall scans prompts, RAG documents, tool argume...
Authenticated Multi Llm Agent

io.github.mikerawsonnz/authenticated-multi-llm-agent

Google-OAuth-gated LLM gateway: verify a Google ID token, then run a Gemini (Vertex AI) completion f