PersonalLeetcodeMCP
Enables Claude to fetch and analyze your public LeetCode submission history, weak areas, and daily challenge via a remote MCP server.
README
LeetCode MCP Connector for Claude.ai
A remote MCP server that gives Claude access to your LeetCode submission history and weak area analysis.
Tools provided
| Tool | Description |
|---|---|
get_submission_history |
Recent submissions with status, language, and multi-attempt detection |
analyze_weak_areas |
Topic-by-topic breakdown of weakest vs strongest areas with recommendations |
get_user_stats |
Overall accepted/submission counts, streaks, and ranking percentile |
get_daily_challenge |
Today's daily challenge problem |
Setup
1. Install & run locally
npm install
npm start
# → Listening on http://localhost:3000
2. Deploy (required for Claude.ai)
Claude.ai requires a publicly accessible HTTPS URL. Choose one:
Option A — Railway (easiest, free tier)
- Push this folder to a GitHub repo
- Go to railway.app → New Project → Deploy from GitHub
- Railway auto-detects Node.js and gives you a public URL like
https://leetcode-mcp-xxx.railway.app
Option B — Render
- Push to GitHub
- render.com → New Web Service → connect repo
- Build:
npm install, Start:node server.js - Gets a URL like
https://leetcode-mcp.onrender.com
Option C — Docker (any VPS/cloud)
docker build -t leetcode-mcp .
docker run -p 3000:3000 leetcode-mcp
# Then point a domain + SSL at port 3000
Option D — Cloudflare Workers (advanced)
Cloudflare has built-in MCP support. See: https://developers.cloudflare.com/mcp/
3. Add to Claude.ai
Once deployed:
- Go to Claude.ai → Settings → Connectors
- Click "Add custom connector"
- Enter your server URL:
https://your-deployment-url.com/mcp - Click Add
That's it — Claude will now have access to the LeetCode tools.
Example prompts
After connecting, try asking Claude:
- "Analyze my weak areas on LeetCode — my username is @john_doe"
- "Show me my last 20 submissions and tell me which problems I struggled with"
- "What are my LeetCode stats and what should I focus on next?"
- "What's today's daily challenge?"
Notes on LeetCode's API
- Public data:
get_submission_history,get_user_stats,analyze_weak_areas, andget_daily_challengeall work with public profile data — no login required, as long as the target profile is public. - Private profiles: If a user has a private profile, LeetCode will return empty data. The user must set their profile to public in LeetCode Settings → Privacy.
- LeetCode does not have an official public API. This server uses the same GraphQL endpoint that LeetCode's own frontend uses. It may break if LeetCode changes their API structure.
Extending this server
To add more tools, follow the pattern in server.js:
server.tool("tool_name", "description", { param: z.string() }, async ({ param }) => {
// call lcQuery() and return results
return { content: [{ type: "text", text: "..." }] };
});
Useful queries to add:
- Problem search by tag/difficulty (uses
problemsetQuestionListquery) - Company-tagged problems (requires LeetCode Premium session cookie)
- Contest history (query is already in the file as
CONTEST_HISTORY_QUERY)
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。