agentsclimarketplace

Mcp builder skills doyajin174 mcp builder

Skill bg-szy/TOP-SKILLS/skills/marketplace/mcp-builder__skills-doyajin174-mcp-builder

Guide for creating MCP (Model Context Protocol) servers. Use this when building integrations with external services, creating new MCP servers, or connecting Claude to APIs.From its SKILL.md

Install
npx -y skills add bg-szy/TOP-SKILLS --skill mcp-builder__skills-doyajin174-mcp-builder

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its file declares

Copied from the file, not written here

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.6 KB, ~1.0k tokens by cl100k_base, as published. Nobody here has run it

MCP Server Builder

MCP 서버를 생성하여 Claude를 외부 서비스와 연결하는 가이드입니다.

Overview

"MCP 서버의 품질은 LLM이 실제 작업을 얼마나 잘 수행할 수 있게 하는가로 측정된다"

Four-Phase Development

Phase 1: Research & Planning

1. MCP 설계 원칙 이해
2. 프로토콜 문서 학습 (mcp.io)
3. 프레임워크 선택 (TypeScript 권장)
4. 대상 API 분석

Phase 2: Implementation

project/
├── src/
│   ├── index.ts      # 진입점
│   ├── client.ts     # API 클라이언트
│   ├── tools/        # 도구 구현
│   └── types.ts      # 타입 정의
├── package.json
└── tsconfig.json

도구 어노테이션:

타입설명
read-only데이터 조회만
destructive데이터 수정/삭제
idempotent반복 실행 안전
open-world외부 시스템 영향

Phase 3: Review & Test

# MCP Inspector로 테스트
npx @anthropic/mcp-inspector

# 도구 호출 테스트
mcp-inspector --server ./dist/index.js

Phase 4: Create Evaluations

10개의 복잡한 평가 질문 생성:

  • 독립적이고 읽기 전용
  • 검증 가능한 답변
  • 실제 사용 시나리오 반영

Technical Recommendations

Transport 선택

환경권장 Transport
로컬 CLIstdio
원격 서버Streamable HTTP
브라우저SSE

TypeScript 구조

import { Server } from "@modelcontextprotocol/sdk/server";
import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio";

const server = new Server({
  name: "my-mcp-server",
  version: "1.0.0",
}, {
  capabilities: {
    tools: {},
  },
});

// 도구 등록
server.setRequestHandler(ListToolsRequestSchema, async () => ({
  tools: [
    {
      name: "get_data",
      description: "데이터 조회",
      inputSchema: {
        type: "object",
        properties: {
          id: { type: "string" }
        },
        required: ["id"]
      }
    }
  ]
}));

// 도구 실행
server.setRequestHandler(CallToolRequestSchema, async (request) => {
  if (request.params.name === "get_data") {
    // 구현
  }
});

// 서버 시작
const transport = new StdioServerTransport();
await server.connect(transport);

Best Practices

도구 네이밍

✅ get_user_profile
✅ create_document
✅ search_files

❌ doThing
❌ process
❌ handle

에러 처리

// 액션 가능한 에러 메시지
throw new Error(
  `API rate limit exceeded. Retry after ${retryAfter} seconds.`
);

// 사용자 가이드 포함
throw new Error(
  `Invalid API key. Get your key at https://api.example.com/keys`
);

상태 관리

  • Stateless JSON 선호 (확장 용이)
  • 세션 상태는 클라이언트에서 관리
  • 캐싱은 선택적으로 구현

Evaluation Format

<evaluation>
  <question>How many active users are in the system?</question>
  <expected_tools>["get_user_count"]</expected_tools>
  <validation>Response contains numeric count</validation>
</evaluation>

References

What ships with it: 1 file

25.5 KB alongside SKILL.md

Keep looking

Skills are one crate of 326,790. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.