SSR/128T MCP Server
Enables AI-powered AIOps management of Juniper SSR/128T networks through natural language, providing 22 tools for router management, alarms, traffic analysis, and more.
README
SSR/128T MCP Server
A Model Context Protocol (MCP) server for the Juniper SSR/128T Networking Platform, enabling AI-powered AIOps management through LLM clients like Claude Desktop and VSCode.
This server provides a bridge between MCP-compatible clients and the SSR Conductor REST API, allowing natural language driven monitoring, visualization, and troubleshooting of your SSR network.
Features
- 🔌 22 MCP Tools covering router management, alarms, traffic analysis, routing tables, interfaces, metrics, and audit logs
- 🔐 Automatic Authentication with token management and auto-refresh
- 🚀 Dual Transport — stdio (local) and streamable-http (remote/deployment)
- 🐳 Docker Support — ready for containerized deployment on any Linux platform
- ⚡ Async HTTP client based on
httpxfor high-performance concurrent requests - 🛡️ Self-Signed Certificate Support for lab and testing environments
Quick Start
Prerequisites
- Python 3.10+ and pip
- Access to an SSR Conductor (v6.x+)
Installation
git clone https://github.com/lotsonlu/ssr-mcp-server.git
cd ssr-mcp-server
pip install -r requirements.txt
Configuration
Set environment variables:
export SSR_BASE_URL="https://<your-conductor-ip>"
export SSR_USERNAME="admin"
export SSR_PASSWORD="your-password"
Or create a config.json file (copy from config-template.json):
{
"conductor": {
"base_url": "https://172.27.69.57",
"username": "admin",
"password": "your-password",
"timeout": 120,
"verify_ssl": false
}
}
Run with Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"ssr-mcp": {
"type": "stdio",
"command": "python3",
"args": ["/path/to/ssr_mcp.py", "-f", "/path/to/config.json", "-t", "stdio"]
}
}
}
Run with VSCode
Add to your .mcp.json:
{
"mcpServers": {
"ssr-mcp": {
"command": "python3",
"args": ["/path/to/ssr_mcp.py", "-f", "/path/to/config.json", "-t", "stdio"]
}
}
}
Run Standalone (HTTP mode)
python ssr_mcp.py -H 0.0.0.0 -p 8080 -t streamable-http
Docker
docker build -t ssr-mcp-server .
docker run --rm -i \
-e SSR_BASE_URL="https://<conductor-ip>" \
-e SSR_USERNAME="admin" \
-e SSR_PASSWORD="your-password" \
ssr-mcp-server -t stdio
Available Tools
Router & System (4)
| Tool | Description |
|---|---|
get_router_list |
List all routers, status, and basic info |
get_router_info |
Detailed info for a specific router |
get_nodes |
List all nodes for a router |
get_system_info |
Conductor system info, version, uptime |
Alarms & Status (4)
| Tool | Description |
|---|---|
get_alarms |
All current alarms (with filters) |
get_router_alarms |
Router-specific alarms |
get_node_alarms |
Node-specific alarms |
get_node_status |
Node operational status & process health |
Traffic & Routing (7)
| Tool | Description |
|---|---|
get_traffic_flows |
Active flow/session table |
get_traffic_fib |
Forwarding Information Base (FIB) |
get_traffic_rib |
Routing Information Base (RIB) |
get_top_sessions |
Top N sessions by bandwidth |
get_top_sources |
Top N traffic sources |
get_adjacencies |
Network adjacencies/peers |
fib_lookup |
FIB lookup by packet headers |
Interfaces & Services (3)
| Tool | Description |
|---|---|
get_device_interfaces |
Physical device interfaces |
get_network_interfaces |
Logical/L3 network interfaces |
get_services |
Configured services & service classes |
Metrics & Analytics (3)
| Tool | Description |
|---|---|
get_router_metrics |
Timeseries performance metrics |
get_analytics |
Aggregated analytics queries |
get_audit_logs |
Historical audit & configuration logs |
Version (1)
| Tool | Description |
|---|---|
get_software_version |
SSR software version per node |
Architecture
┌──────────────────────────────────────────────┐
│ Claude Desktop / VSCode / Other MCP Clients │
└──────────────────┬───────────────────────────┘
│ stdio / streamable-http
┌──────────────────▼───────────────────────────┐
│ SSR MCP Server (ssr_mcp_server) │
│ ├── server.py MCP Server & Tool Defs │
│ ├── ssr_client.py REST API Client │
│ ├── tools/ 22 Tool Handlers │
│ └── utils/ Config Management │
└──────────────────┬───────────────────────────┘
│ REST API (HTTPS)
┌──────────────────▼───────────────────────────┐
│ SSR Conductor │
│ /router, /alarm, /traffic, /stats, etc. │
└──────────────────────────────────────────────┘
Project Structure
ssr-mcp-server/
├── ssr_mcp.py # CLI entry point
├── ssr_mcp_server/ # Core package
│ ├── server.py # MCP server & 22 tool definitions
│ ├── ssr_client.py # REST API client (auth, HTTP, retry)
│ ├── tools/ # Tool handler modules
│ │ ├── router_tools.py # Router & system tools
│ │ ├── alarm_tools.py # Alarm & status tools
│ │ ├── traffic_tools.py # Traffic & routing tools
│ │ ├── interface_tools.py # Interface & service tools
│ │ ├── metrics_tools.py # Metrics & analytics tools
│ │ └── audit_tools.py # Audit & version tools
│ └── utils/
│ └── config.py # Configuration management
├── pyproject.toml # Project metadata & dependencies
├── requirements.txt # Pip dependencies
├── Dockerfile # Docker build
├── Makefile # Dev/build helpers
├── config-template.json # Configuration template
└── LICENSE
Environment Variables
| Variable | Default | Description |
|---|---|---|
SSR_BASE_URL |
https://127.0.0.1 |
SSR Conductor URL |
SSR_USERNAME |
admin |
API username |
SSR_PASSWORD |
— | API password (required) |
SSR_TIMEOUT |
120 |
API timeout in seconds |
SSR_VERIFY_SSL |
false |
Enable SSL verification |
Development
# Install dependencies
pip install -r requirements.txt
# Run with stdio transport
python ssr_mcp.py -t stdio
# Run tests
python -m pytest tests/
# Build Docker image
docker build -t ssr-mcp-server .
License
Acknowledgments
This project is inspired by and follows patterns from the official Juniper Junos MCP Server.
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。