Higress OPS MCP Server
by higress-group
This repository provides a Model Context Protocol (MCP) server implementation for comprehensive configuration and management of Higress. It also includes an MCP client built on LangGraph and LangChain MCP Adapters, facilitating interaction with the server through an agent flow architecture.
Last updated: N/A
What is Higress OPS MCP Server?
The Higress OPS MCP Server is a Model Context Protocol server that enables comprehensive configuration and management of Higress. It also includes an MCP client for interacting with the server.
How to use Higress OPS MCP Server?
To use the server, configure the environment variables, start the MCP client (which also starts the server), and add new tools by creating a tool class, adding a method to HigressClient (if needed), registering the tool class in the server, and adding the tool to SENSITIVE_TOOLS if it requires human confirmation.
Key features of Higress OPS MCP Server
Comprehensive Higress configuration management
MCP server implementation
MCP client with LangGraph and LangChain
Agent flow architecture
Extensible tool framework
Human confirmation for sensitive operations
Use cases of Higress OPS MCP Server
Automated Higress configuration
Centralized management of Higress resources
Integration with LangChain-based agents
Custom tool development for specific Higress tasks
Secure configuration changes with human oversight
FAQ from Higress OPS MCP Server
What is MCP?
What is MCP?
MCP stands for Model Context Protocol, a protocol for distributing configuration data.
What is Higress?
What is Higress?
Higress is a cloud-native gateway.
What are LangGraph and LangChain MCP Adapters?
What are LangGraph and LangChain MCP Adapters?
LangGraph and LangChain MCP Adapters are used to build the MCP client and facilitate interaction with the Higress MCP Server.
How do I add a new tool?
How do I add a new tool?
Create a new tool class, add a method to HigressClient (if needed), register the tool class in the server, and add the tool to SENSITIVE_TOOLS if it requires human confirmation.
Why are some tools marked as sensitive?
Why are some tools marked as sensitive?
Sensitive tools require human confirmation before execution to prevent unintended or malicious changes.