Cisco Meraki MCP Server
Provides MCP tools for Cisco Meraki Dashboard API, enabling management of networks, configuration templates, and health checks via natural language commands.
README
Cisco Meraki MCP Server
Overview
The Cisco Meraki MCP Server is a Python-based implementation of the Model Context Protocol (MCP) that exposes Cisco Meraki Dashboard API operations as MCP tools.
Instead of calling REST APIs directly, any MCP-compatible client can invoke these tools using a standardized interface.
This project demonstrates how MCP can be used to bridge AI assistants and enterprise networking platforms.
Features
- Cisco Meraki Dashboard API integration
- Model Context Protocol (MCP) Server
- Streamable HTTP transport
- Environment variable support
- Interactive MCP client
- Clean and modular Python code
- Production-ready project structure
- Easy to extend with additional Meraki APIs
Supported Tools
Health
- Health Check
Networks
- Create Network
- Delete Network
- List Networks
- Get Network Information
Configuration Templates
- Create Template
- List Templates
- Get Template Information
- Bind Template to Network
Project Structure
meraki-mcp/
│
├── app.py # MCP Server
├── client.py # Interactive MCP Client
├── meraki.py # Cisco Meraki REST API Wrapper
├── config.py # Configuration & Environment Variables
├── requirements.txt
├── README.md
├── .env
Requirements
- Python 3.10 or later
- Cisco Meraki Dashboard Account
- Cisco Meraki API Key
Installation
Clone the repository.
git clone <repository-url>
Move into the project directory.
cd meraki-mcp
Install the required packages.
pip install -r requirements.txt
Environment Variables
Create a .env file in the project directory (or copy .env.example to .env).
Example:
# Cisco Meraki Dashboard API Key
MERAKI_API_KEY=YOUR_MERAKI_API_KEY
# OpenAI API Key
# Required only if using the AI-powered client.
OPENAI_API_KEY=YOUR_OPENAI_API_KEY
Note:
MERAKI_API_KEYis required to authenticate with the Cisco Meraki Dashboard API.OPENAI_API_KEYis only required if you are using the AI-powered client that converts natural language into MCP tool calls.
Running the MCP Server
Start the server.
python app.py
If successful, you should see output similar to:
Connected to Cisco Meraki Dashboard
Organization ID : xxxxxxx
Cisco Meraki MCP Server Started
Running the Interactive Client
Open another terminal.
Run:
python client.py
Example:
============================================================
Cisco Meraki Natural Language MCP Client
============================================================
Connected Successfully.
You >
Supported Natural Language Commands
Health
Health check
Check health
Check server health
Server status
Status
Networks
Show all networks
List all networks
List networks
Create a network named Delhi Office
Create network Delhi Office
Delete network Delhi Office
Remove network Delhi Office
Show network Delhi Office
Get network Delhi Office
Show information for Delhi Office
Templates
List templates
Show templates
Create a template named Branch Template
Create template Branch Template
Show template Branch Template
Get template Branch Template
Bind template Branch Template to Delhi Office
Apply template Branch Template to Delhi Office
Other Commands
Help
Exit
Example Session
You > Show all networks
Result
------------------------------------------------------------
[
{
"id": "...",
"name": "Delhi Office"
}
]
You > Create a network named Mumbai Office
Result
------------------------------------------------------------
Network created successfully.
You > List templates
Result
------------------------------------------------------------
...
You > Exit
Current Architecture
+-----------------------------+
| Interactive Client |
| (client.py) |
+-------------+---------------+
|
|
v
+-----------------------------+
| MCP Server |
| (app.py) |
+-------------+---------------+
|
|
v
+-----------------------------+
| Cisco Meraki REST Wrapper |
| (meraki.py) |
+-------------+---------------+
|
|
v
+-----------------------------+
| Cisco Meraki Dashboard API |
+-----------------------------+
Technology Stack
- Python
- Model Context Protocol (MCP)
- FastMCP
- Cisco Meraki Dashboard API
- Requests
- Python Dotenv
- Uvicorn
Future Enhancements
- VLAN Management
- Switch Port Configuration
- Wireless SSID Management
- Firewall Rules
- Device Inventory
- Client Information
- Organization Administration
- OpenAI Integration for intelligent natural language understanding
- Multi-Organization Support
- Logging
- Unit Testing
- Docker Support
- GitHub Actions CI/CD
Troubleshooting
Invalid API Key
Ensure the .env file contains a valid Cisco Meraki API key.
Network Not Found
Verify that the specified network exists in your Cisco Meraki Dashboard.
Template Not Found
Verify that the template exists before attempting to retrieve or bind it.
Connection Errors
Ensure the MCP server is running before starting the client.
Learning Objectives
This project demonstrates:
- Building an MCP Server from scratch
- Creating custom MCP tools
- Calling enterprise REST APIs
- Environment variable management
- Python project organization
- Interactive MCP client development
- Cisco Meraki Dashboard API automation
Author
Diwakar Sharma
Python Developer | Network Automation | AI | Cisco Meraki | MCP | REST APIs
License
This project is provided for educational purposes. Modify and extend it according to your own requirements.
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。