juniper-mist-mcp

juniper-mist-mcp

MCP server for managing Juniper Mist networks with AI assistants. Monitor, troubleshoot, and configure customer networks through natural language.

Category
访问服务器

README

Juniper Mist MCP Server

⚠️ Juniper now offers an official MCP server for Mist.

Juniper has released their own MCP server at https://mcp.ai.juniper.net/mcp/mist with native integration for Claude Desktop.

We recommend using the official upstream server instead of this project. See the Juniper Mist MCP documentation for setup instructions.

This community project remains available for use cases that require multi-org MSP routing, write tools, or HTTP transport — features not yet covered by the upstream server. However, for standard single-org monitoring and troubleshooting, the official server is the better choice going forward.

MCP server for managing Juniper Mist networks with AI assistants. Monitor, troubleshoot, and configure customer networks through natural language — built for MSPs managing multiple organizations across regions.

Input: Natural language via Claude Desktop (or any MCP client) Output: Device stats, client data, alarms, WLAN configs, inventory, and more

  AI Assistant  ──▶  MCP Server  ──▶  Juniper Mist API
  (Claude, etc.)     (multi-org       (US, EU, GC1, GC2,
                      routing)         APAC regions)

Installation

Option 1: Install from PyPI (recommended)

pip install juniper-mist-mcp

Option 2: Install with pipx (isolated environment)

pipx install juniper-mist-mcp

Verify installation

juniper-mist-mcp --help

Setup

1. Get a Mist API Token

  1. Log in to the Mist portal at https://admin.mist.com
  2. Navigate to Organization → Settings → API Tokens
  3. Create a new API token with appropriate scope

2. Configure Organizations

Create a .env file with your organization credentials:

# US-based org (default region)
MIST_TOKEN_acme_corp=your_token_here

# EU-based org
MIST_TOKEN_euro_client=your_token_here
MIST_REGION_euro_client=api.eu.mist.com

# Government org
MIST_TOKEN_gov_agency=your_token_here
MIST_REGION_gov_agency=api.gc1.mist.com

Each org needs a MIST_TOKEN_<ORGNAME> variable. Region is optional — defaults to US (api.mist.com).

Supported Regions

Region Endpoint
US (default) api.mist.com
EU api.eu.mist.com
Government Cloud 1 api.gc1.mist.com
Government Cloud 2 api.gc2.mist.com
Asia Pacific api.ac2.mist.com

Usage

Run in stdio mode (Claude Desktop)

# Read-only mode (default — 10 tools)
juniper-mist-mcp

# With write tools enabled (14 tools)
juniper-mist-mcp --enable-write-tools

Run in HTTP mode (centralized deployment)

# Default port 8000
juniper-mist-mcp --transport=http

# Custom host and port
juniper-mist-mcp --transport=http --host=0.0.0.0 --port=8080

# HTTP + write tools
juniper-mist-mcp --transport=http --port=8080 --enable-write-tools

Configure Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS):

{
  "mcpServers": {
    "juniper-mist": {
      "command": "juniper-mist-mcp",
      "env": {
        "MIST_TOKEN_acme_corp": "your_token_here",
        "MIST_REGION_acme_corp": "api.mist.com"
      }
    }
  }
}

Or if using a .env file alongside the install:

{
  "mcpServers": {
    "juniper-mist": {
      "command": "juniper-mist-mcp",
      "args": ["--env-file", "/path/to/your/.env"]
    }
  }
}

Verify it works

In your AI assistant, ask it to list organizations. The mist_list_orgs tool should return your configured orgs:

[
  {"name": "acme_corp", "region": "api.mist.com", "has_token": true},
  {"name": "euro_client", "region": "api.eu.mist.com", "has_token": true}
]

Available Tools

Read Tools (always available)

Tool What it does
mist_list_orgs List configured customer orgs and regions
mist_get_device_stats AP, switch, and gateway statistics
mist_get_sle_summary Service Level Experience metrics (throughput, latency, coverage)
mist_get_client_stats Wireless client connection and bandwidth data
mist_get_alarms Infrastructure, security, and Marvis AI alarms
mist_get_site_events Config changes, user activities, system alerts
mist_list_wlans WLAN/SSID profiles and security settings
mist_get_rf_templates Radio frequency templates (channel, power, band)
mist_get_inventory Device inventory with type/status/site filters
mist_get_device_config_cmd Generated CLI config commands for a device

Write Tools (opt-in with --enable-write-tools)

Tool What it does
mist_update_wlan Modify WLAN/SSID configurations
mist_manage_nac_rules Create, update, or delete 802.1X NAC rules
mist_manage_wxlan Manage WXLAN microsegmentation policies
mist_manage_security_policies Create, update, or delete security policies

Every tool takes an org parameter to route to the right customer org and region.


Safety Model

Four layers prevent accidental network changes:

  1. Read-only by default — write tools aren't registered unless you opt in
  2. CLI flag — must explicitly pass --enable-write-tools
  3. MCP annotations — read tools tagged readOnlyHint, write tools tagged destructiveHint so the AI asks for confirmation
  4. Pre-flight validation — UUID format checks and required field validation before any API call

CLI Reference

Flag Description Default
--enable-write-tools Enable 4 write tools for network modification Disabled
--transport Transport protocol: stdio or http stdio
--host Host for HTTP transport 0.0.0.0
--port Port for HTTP transport 8000
--env-file Path to .env file .env

Troubleshooting

Error Fix
Configuration file not found: .env Create a .env file from .env.example
Organization 'X' is not configured Check MIST_TOKEN_<ORGNAME> in your .env (case-sensitive)
Unknown region 'X' Use a supported region endpoint (see table above)
Failed to fetch orgs: 401 Token is invalid or expired — regenerate in Mist portal
Write tools not appearing Start server with --enable-write-tools flag

Contributing

Build from source

git clone https://github.com/vahagn-madatyan/juniper-mist-mcp.git
cd juniper-mist-mcp

# Create virtual environment
python -m venv .venv
source .venv/bin/activate  # or .venv\Scripts\activate on Windows

# Install with dev dependencies
pip install -e ".[dev]"

# Run from source
python -m mist_mcp.server --help

Run tests

pytest tests/ -v
pytest tests/ -v --tb=short
pytest tests/test_config.py -v

Project structure

mist_mcp/
├── __init__.py        # Package version
├── __main__.py        # Module entry point
├── server.py          # MCP server, tool definitions, CLI
├── config.py          # .env loader, org routing, region validation
└── session.py         # Mist API session management

Architecture

  • FastMCP — MCP server framework handling tool registration, lifespan, and transport
  • mistapi SDK — Juniper Mist Python SDK for all API calls
  • Multi-tenant routing — each tool call includes an org parameter; the server maintains separate authenticated sessions per org
  • Conditional tool registration — write tools are only registered when --enable-write-tools is passed, using mcp.add_tool() with MCP annotations

Making a release

Releases are published to PyPI automatically via GitHub Actions when a version tag is pushed:

# Update version in pyproject.toml and mist_mcp/__init__.py
# Commit the version bump
git tag v0.2.0
git push origin v0.2.0

The CI pipeline builds the distribution, runs tests, publishes to TestPyPI on every push, and publishes to PyPI on tags matching v*.


License

Apache License 2.0

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选