Find Rules MCP Server

❌ Without Find Rules

LLMs often struggle with maintaining consistent coding standards and best practices:

• ❌ Inconsistent code quality and standards across different tasks
• ❌ Missing critical technical considerations and edge cases
• ❌ Overlooked best practices and security requirements

✅ With Find Rules

Find Rules MCP analyzes your task requirements and provides comprehensive guidance before you start coding:

• ✅ Clear, task-specific guidance rules
• ✅ Best practices tailored to your tech stack
• ✅ Quality standards and technical considerations

Add use find_rules to your prompt in Trae/Cursor/Windsurf:

Create a basic Next.js project with app router. use find_rules

Create a script to delete the rows where the city is "" given PostgreSQL credentials. use find_rules

Get instant guidance:

• 1️⃣ Describe your task requirements naturally
• 2️⃣ Tell the LLM to use find_rules
• 3️⃣ Receive comprehensive guidelines

No more inconsistent code quality, no missed best practices, no overlooked technical considerations.

Technical Details

Tool Description

The find_rules tool serves as an essential pre-task step to ensure task execution follows best practices and standard specifications. It analyzes user requirements and technical context to provide the most appropriate guidance rules.

Parameters

• user_input: Original task requirements from the user (string type, required)
• tags: Related technology tags (one or more) (string array type, required)

Return Value

The tool returns a response containing guidance rules in the following format:

{
  "content": [
    {
      "type": "text",
      "text": "Please follow these guidelines to complete your task:\n\n1. [Task-specific guidance rules]\n2. [Best practice recommendations]\n3. [Quality standard requirements]\n..."
    }
  ]
}

Configuration

Environment Variables

This service supports configuration through environment variables:

OPENAI_API_KEY

• Description: OpenAI API key for calling large model services
• Importance: Configuring this variable can significantly improve rule selection accuracy
• Default: None, strongly recommended to configure this variable

OPENAI_BASE_URL

• Description: Base URL for OpenAI API
• Default: https://openrouter.ai/api/v1
• Usage: Can be configured to use other compatible API services

OPENAI_MODEL

• Description: Name of the large model to use
• Default: qwen/qwen3-8b:free
• Usage: Can choose different models to balance performance and cost

Integration

MCP Client Integration

This server can be integrated with any client that supports the Model Context Protocol (MCP). Here's how to set it up:

Claude Desktop Configuration

{
  "mcpServers": {
    "Guidance Rules Finder": {
      "command": "npx",
      "args": ["-y", "find-rules-server"],
      "env": {
        "OPENAI_API_KEY": "your-openrouter-api-key"
      }
    }
  }
}

Other MCP Clients

For other MCP clients, you'll need:

• Server start command: find-rules-server
• Server name: Custom identifier name

Refer to your client's documentation for specific configuration details.

Development

For development guidelines and rule customization, see Development Guide.

Credits

All rule data is sourced from Cursor Directory. Copyright belongs to the original authors.

License

MIT