leetcode-helper
An MCP server that provides structured, AI-powered guidance for LeetCode problems, including approach analysis, hints, complexity breakdowns, and optimization directions without spoiling solutions.
README
leetcode-helper
An MCP server that gives structured, AI-powered DSA guidance inside Claude Desktop or Claude Code — approach analysis, hints, complexity breakdowns, and more, without spoiling the solution.
Tools at a Glance
| Tool | Input | What it returns |
|---|---|---|
get_approach |
problem | Pattern, steps, TC/SC |
get_hint |
problem + where stuck | Nudge, no spoilers |
get_complexity |
problem + code | TC/SC breakdown per block |
post_solution |
problem + accepted code | Discussion post ready to copy |
similar_questions |
problem | 3–5 problems for the same pattern |
upsolve |
problem + working code | Optimization direction, not solution |
Setup
Prerequisites: Node.js 18+, an ANTHROPIC_API_KEY
git clone https://github.com/Navoren/leetcode-helper.git
cd leetcode-helper
npm install
npm run build
Connect to Claude Code
Add to .claude/settings.json (project or global):
{
"mcpServers": {
"leetcode-helper": {
"command": "node",
"args": ["d:/leetcode-helper/dist/index.js"],
"env": {
"ANTHROPIC_API_KEY": "your-key-here"
}
}
}
}
Connect to Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"leetcode-helper": {
"command": "node",
"args": ["/absolute/path/to/leetcode-helper/dist/index.js"],
"env": {
"ANTHROPIC_API_KEY": "your-key-here"
}
}
}
}
Example Usage
Getting an approach:
"Use get_approach on: Given an array of integers, return indices of two numbers that add up to target."
Getting a hint when stuck:
"Use get_hint — problem: two sum, stuck on: my nested loop works but is too slow"
Generating a discussion post:
"Use post_solution with my accepted Python solution for two sum"
Upsolving:
"Use upsolve — here's my O(n²) solution, help me think toward better"
Project Structure
leetcode-mcp/
├── src/
│ ├── index.ts # Entry point
│ ├── server.ts # McpServer setup
│ ├── lib/
│ │ └── claude.ts # Anthropic API client
│ └── tools/
│ ├── registry.ts # Registers all tools (one line per tool)
│ ├── get_approach.ts
│ ├── get_hint.ts
│ ├── get_complexity.ts
│ ├── post_solution.ts
│ ├── similar_questions.ts
│ └── upsolve.ts
├── dist/ # Compiled output
├── functional_requirement.md
├── technical_document.md
└── CLAUDE.md
Adding a New Tool
- Create
src/tools/your_tool.ts— exportname,description,schema(Zod),handler - Add one import + one entry in
src/tools/registry.ts npm run build
No changes to server.ts, index.ts, or claude.ts.
Tech Stack
| Layer | Choice |
|---|---|
| Runtime | Node.js (ESM) |
| Language | TypeScript 5 |
| MCP SDK | @modelcontextprotocol/sdk |
| Schema validation | Zod |
| LLM | Anthropic claude-sonnet-4-6 |
| Transport | stdio |
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。