Higress OPS MCP Server

1# Higress OPS MCP Server
2 
3A Model Context Protocol (MCP) server implementation that enables comprehensive configuration and management of [Higress](https://higress.cn/). This repository also provides an MCP client built on top of [LangGraph](https://www.langchain.com/langgraph) and [LangChain MCP Adapters](https://github.com/langchain-ai/langchain-mcp-adapters), facilitating interaction with the Higress MCP Server through a well-designed agent flow architecture.
4 
5## Demo
6 
7https://github.com/user-attachments/assets/bae66b77-a158-452e-9196-98060bac0df7
8 
9## Config Environment Variables
10 
11Copy the `.env.example` file to `.env` and fill in the corresponding values.
12 
13## Start MCP Client and MCP Server
14 
15In stdio mode, the MCP server process is started by the MCP client program. Run the following command to start the MCP client and MCP server:
16 
17```python
18uv run client.py
19```
20 
21## Add a new tool
22 
23**Step 1: Create a new tool class or extend an existing one**
24 
25- Create a new file in the tools directory if adding a completely new tool category
26- Or add your tool to an existing class if it fits an existing category
27 
28```python
29from typing import Dict, List, Any
30from fastmcp import FastMCP
31 
32class YourTools:
33    def register_tools(self, mcp: FastMCP):
34        @mcp.tool()
35        async def your_tool_function(arg1: str, arg2: int) -> List[Dict]:
36            """
37            Your tool description.
38            
39            Args:
40                arg1: Description of arg1
41                arg2: Description of arg2
42 
43            Returns:
44                Description of the return value
45            
46            Raises:
47                ValueError: If the request fails
48            """
49            # Implementation using self.higress_client to make API calls
50            return self.higress_client.your_api_method(arg1, arg2)
51```
52 
53 
54**Step 2: Add a new method to HigressClient if your tool needs to interact with the Higress Console API**
55 
56- Add methods to utils/higress_client.py that encapsulate API calls
57- Use the existing HTTP methods (get, put, post) for actual API communication
58 
59 
60```python
61def your_api_method(self, arg1: str, arg2: int) -> List[Dict]:
62    """
63    Description of what this API method does.
64    
65    Args:
66        arg1: Description of arg1
67        arg2: Description of arg2
68        
69    Returns:
70        Response data
71        
72    Raises:
73        ValueError: If the request fails
74    """
75    path = "/v1/your/api/endpoint"
76    data = {"arg1": arg1, "arg2": arg2}
77    return self.put(path, data)  # or self.get(path) or self.post(path, data)
78```
79 
80**Step 3: Register your tool class in the server**
81 
82- Add your tool class to the tool_classes list in server.py
83- This list is used by ToolsRegister to instantiate and register all tools
84- The ToolsRegister will automatically set logger and higress_client attributes
85 
86```python
87tool_classes = [
88    CommonTools,
89    RequestBlockTools,
90    RouteTools,
91    ServiceSourceTools,
92    YourTools  # Add your tool class here
93]
94```
95 
96**Step 4: Add your tool to `SENSITIVE_TOOLS` if it requires human confirmation**
97 
98- Tools in this list will require human confirmation before execution
99 
100```python
101# Define write operations that require human confirmation
102SENSITIVE_TOOLS = [
103    "add_route", 
104    "add_service_source",
105    "update_route",
106    "update_request_block_plugin", 
107    "update_service_source",
108    "your_tool_function"  # Add your tool name here if it requires confirmation
109]
110```
111