Inky MCP Server
MCP server that connects Cursor IDE to Inky backend, enabling users to list and filter personalized coding rules stored in Pinecone.
README
Inky MCP Server
MCP (Model Context Protocol) server for Inky that provides access to your personalized coding rules stored in Pinecone. This server acts as a bridge between Cursor IDE and your Inky backend API.
📦 Published on npmjs: https://www.npmjs.com/package/inky-mcp-server
Overview
The Inky MCP server allows you to access your personalized software rules directly from Cursor IDE. It connects to your hosted Inky API endpoint and retrieves rules from your Pinecone database.
Features
- ✅ List all your coding rules
- ✅ Filter rules by repository
- ✅ Secure API key authentication
- ✅ JSON-RPC 2.0 protocol compliance
- ✅ Compatible with Cursor IDE
Installation
The package is available on npmjs: inky-mcp-server
Using npx (Recommended)
npx -y inky-mcp-server
Using npm
npm install -g inky-mcp-server
Using pnpm
pnpm add -g inky-mcp-server
Using yarn
yarn global add inky-mcp-server
Configuration
1. Get Your API Token
First, you need to generate an MCP access token from your Inky account. The token will be in the format inky_....
2. Configure Cursor IDE
Add the following configuration to your Cursor settings. The configuration file location depends on your OS:
macOS/Linux: ~/.cursor/mcp.json
Windows: %APPDATA%\Cursor\mcp.json
Option 1: HTTP Transport (Recommended - Direct API Access)
{
"mcpServers": {
"inky": {
"url": "https://your-domain.com/api/mcp",
"headers": {
"Authorization": "Bearer inky_YOUR_TOKEN_HERE"
}
}
}
}
Option 2: Stdio Transport (Using MCP Server Package)
{
"mcpServers": {
"inky": {
"command": "npx",
"args": ["-y", "inky-mcp-server"],
"env": {
"API_KEY": "inky_YOUR_TOKEN_HERE",
"INKY_API_URL": "https://your-domain.com"
}
}
}
}
Environment Variables
API_KEY(required): Your Inky MCP access token (starts withinky_)INKY_API_URL(optional): Your Inky API base URL. Defaults to:NEXT_PUBLIC_APP_URLenvironment variablehttp://localhost:3000if not set
Usage
Once configured, the MCP server will be available in Cursor IDE. You can use it to:
List All Rules
Query all your coding rules:
List all my coding rules
Filter by Repository
Get rules for a specific repository:
Show me rules for repository [repository_id]
API Reference
The MCP server communicates with the Inky backend API using JSON-RPC 2.0 protocol.
Tools
list_rules
Lists all user's rules, optionally filtered by repository.
Parameters:
repository_id(optional): Filter rules by repository ID
Example Request:
{
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"name": "list_rules",
"arguments": {
"repository_id": "repo_123"
}
},
"id": 1
}
Example Response:
{
"jsonrpc": "2.0",
"result": {
"content": [
{
"type": "text",
"text": "[{\"id\":\"rule_123\",\"name\":\"TypeScript Best Practices\",\"content\":\"...\",\"version\":1,\"is_active\":true,\"repository_id\":\"repo_123\",\"created_at\":\"2024-01-01T00:00:00Z\"}]"
}
]
},
"id": 1
}
Development
Building from Source
cd packages/mcp-server
pnpm install
pnpm build
Running in Development Mode
pnpm dev
Testing
You can test the MCP server using the Postman collection included in the main repository (postman-mcp-collection.json).
Architecture
┌─────────────┐
│ Cursor IDE │
└──────┬──────┘
│ stdio
│
┌──────▼──────────────────┐
│ Inky MCP Server │
│ (This Package) │
└──────┬──────────────────┘
│ HTTP POST
│ Bearer Token
│
┌──────▼──────────────────┐
│ Inky Backend API │
│ /api/mcp │
└──────┬──────────────────┘
│
┌──────▼──────────────────┐
│ Pinecone Database │
│ (User's Rules) │
└──────────────────────────┘
Security
- API tokens are passed via environment variables (never hardcoded)
- All API requests use Bearer token authentication
- Tokens are hashed and stored securely in the database
- Each user can only access their own rules
Troubleshooting
Server Not Starting
Error: Error: API_KEY environment variable is required
Solution: Make sure you've set the API_KEY environment variable in your Cursor MCP configuration.
Connection Errors
Error: Backend API error
Solution:
- Verify your
INKY_API_URLis correct - Check that your API token is valid
- Ensure the backend API is accessible
No Rules Returned
Solution:
- Verify you have rules created in your Inky account
- Check that your API token belongs to the correct user
- Try listing rules without repository filter first
License
MIT
Support
For issues and questions, please refer to the main Inky repository.
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。