deepseek-mcp-server MCP — 사용 사례, 설치 & 라이브 데모

왜 쓰나요

핵심 기능

4 tools: chat, completions, models, balance
Code execution mode — model writes and runs a program that calls tools
Remote endpoint (https://deepseek-mcp.ragweld.com/mcp) + local npm + Docker
Bearer token for remote; DEEPSEEK_API_KEY for local
Works from Cursor, Codex, Claude Code

라이브 데모

실제 사용 모습

deepseek.replay ▶ 준비됨

0/0

설치

클라이언트 선택

~/Library/Application Support/Claude/claude_desktop_config.json · Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "deepseek": {
      "command": "npx",
      "args": [
        "-y",
        "deepseek-mcp-server"
      ],
      "_inferred": true
    }
  }
}

Claude Desktop → Settings → Developer → Edit Config 열기. 저장 후 앱 재시작.

~/.cursor/mcp.json · .cursor/mcp.json

{
  "mcpServers": {
    "deepseek": {
      "command": "npx",
      "args": [
        "-y",
        "deepseek-mcp-server"
      ],
      "_inferred": true
    }
  }
}

Cursor는 Claude Desktop과 동일한 mcpServers 스키마 사용. 프로젝트 설정이 전역보다 우선.

VS Code → Cline → MCP Servers → Edit

{
  "mcpServers": {
    "deepseek": {
      "command": "npx",
      "args": [
        "-y",
        "deepseek-mcp-server"
      ],
      "_inferred": true
    }
  }
}

Cline 사이드바의 MCP Servers 아이콘 클릭 후 "Edit Configuration" 선택.

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "deepseek": {
      "command": "npx",
      "args": [
        "-y",
        "deepseek-mcp-server"
      ],
      "_inferred": true
    }
  }
}

Claude Desktop과 같은 형식. Windsurf 재시작 후 적용.

~/.continue/config.json

{
  "mcpServers": [
    {
      "name": "deepseek",
      "command": "npx",
      "args": [
        "-y",
        "deepseek-mcp-server"
      ]
    }
  ]
}

Continue는 맵이 아닌 서버 오브젝트 배열 사용.

~/.config/zed/settings.json

{
  "context_servers": {
    "deepseek": {
      "command": {
        "path": "npx",
        "args": [
          "-y",
          "deepseek-mcp-server"
        ]
      }
    }
  }
}

context_servers에 추가. 저장 시 Zed가 핫 리로드.

claude mcp add deepseek -- npx -y deepseek-mcp-server

한 줄 명령. claude mcp list로 확인, claude mcp remove로 제거.

사용 사례

실전 활용법: deepseek-mcp-server

Delegate cheap-but-long reasoning to DeepSeek from inside Claude

👤 Cost-conscious devs building agent flows ⏱ ~10 min intermediate

언제 쓸까: You have a step that needs chain-of-thought but not Claude-level quality — save 90% on tokens.

사전 조건

DeepSeek API key — platform.deepseek.com — cheap

흐름

Route specific subtask

Use deepseek chat (model deepseek-reasoner) to classify these 100 support tickets into 5 buckets. Return only the classifications.✓ 복사됨

→ Classifications completed cheaply

결과: Same quality for classification/extraction at a fraction of the cost.

함정

Quality gap on nuanced writing tasks — Profile first — DeepSeek shines on structured tasks, trails on creative/long-form

Let DeepSeek write a short script to call multiple tools in one go

👤 Agent builders tired of 10-round tool call loops ⏱ ~20 min advanced

언제 쓸까: A task needs 8 tool calls in sequence; going round-trip each time is slow and burns tokens.

흐름

Enable code-exec mode

Using deepseek code-exec mode: fetch 5 URLs with fetch MCP, summarize each, then combine into a table. One shot.✓ 복사됨

→ Script runs inside exec runtime; returns table

결과: Fewer tokens, fewer round-trips for pipeline-style tasks.

함정

Script has bugs in the execution runtime — Fall back to per-tool calls when debugging
Security — exec runtime runs LLM-authored code — Keep tool access narrow; don't grant filesystem/shell unless essential

함께 쓰기: fetch · firecrawl

Monitor DeepSeek API spend from your agent

👤 Ops people running DeepSeek-backed agents ⏱ ~5 min beginner

언제 쓸까: You want an agent to halt when spend is low or to alert weekly.

흐름

Pull balance

What's my DeepSeek balance?✓ 복사됨

→ Current balance
Alert if low

If balance < $5, post a Slack message #ops-alerts with the number.✓ 복사됨

→ Conditional alert

결과: Never run out of credits mid-pipeline.

조합

다른 MCP와 조합해 10배 효율

deepseek + fetch

Scrape then classify with DeepSeek cheaply

Fetch 20 news URLs, then deepseek-classify each by topic (5 buckets).✓ 복사됨

도구

이 MCP가 노출하는 것

도구	입력	언제 호출	비용
chat	messages: Msg[], model?: str, stream?: bool	Delegate a reasoning or classification task	DeepSeek tokens
completions	prompt, model?, max_tokens?	Non-chat completion calls	DeepSeek tokens
models		List available DeepSeek models	free
balance		Check spend / alert on low credit	free

비용 및 제한

운영 비용

API 쿼터: Per DeepSeek account
호출당 토큰: DeepSeek is 10-50x cheaper than frontier models for many tasks
금액: DeepSeek V3 ~$0.27/M input, R1 ~$0.55/M input (as of 2026)
팁: Use the balance tool to hard-stop agents when credits drop below your threshold.

보안

권한, 시크릿, 파급범위

자격 증명 저장: DEEPSEEK_API_KEY for local; DEEPSEEK_MCP_AUTH_TOKEN for remote endpoint

데이터 외부 송신: Your prompts go to api.deepseek.com (China-based)

절대 부여 금지: Wide tool surface to the code-exec mode without reviewing which tools are exposed

DeepSeek stores inputs; read their data policy. Avoid sending confidential content.
Code execution mode runs LLM-authored code — treat with same caution as any agent code-exec sandbox.

문제 해결

자주 발생하는 오류와 해결

401 from DeepSeek

Check API key. Also verify account hasn't been suspended.

확인: curl -H 'Authorization: Bearer $KEY' https://api.deepseek.com/v1/models

Rate limit

DeepSeek has tight per-minute limits on new accounts. Add retry + backoff.

Code-exec mode returns 'tool not available'

Code-exec sees only the tool set you exposed. Add the needed tool to its allowlist.

대안

deepseek-mcp-server 다른 것과 비교

대안	언제 쓰나	단점/장점
OpenAI / Anthropic via direct API	You want frontier quality and are not cost-bound	5-50x the cost
Groq MCP	You want ultra-low latency on open models	Different model lineup; no code-exec mode

deepseek-mcp-server