Skills Plugins MCP Prompt Model 博客 我的中心
生活与工具 #api #ai #mcp

zabbix-mcp-server

Complete MCP server for Zabbix integration providing AI assistants access to 100% of Zabbix APIs through 3 unified tools

DeepseekModel 官方收录技能 质量 良好 · 64 v1.0.0

获取

https://deepseekmodel.com/api/download.php?id=aradotso-mcp-skills-skills-zabbix-mcp-server-skill-md&format=skill
下载 .skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用
.skill 文件中 system_prompt 字段的实际内容。
name zabbix-mcp-server description Complete MCP server for Zabbix integration providing AI assistants access to 100% of Zabbix APIs through 3 unified tools triggers ["connect to zabbix monitoring","query zabbix hosts and problems","integrate zabbix with AI assistant","access zabbix api through mcp","monitor infrastructure with zabbix","get zabbix metrics and alerts","configure zabbix mcp server","read zabbix monitoring data"] Zabbix MCP Server Skill by ara.so — MCP Skills collection. A lightweight Model Context Protocol (MCP) server that provides complete access to the entire Zabbix API (100+ methods) through just 3 tools. Compatible with Zabbix 6.0+. What It Does The Zabbix MCP Server bridges AI assistants with Zabbix monitoring infrastructure, enabling: Query hosts, items, triggers, templates, problems, and metrics Create and manage monitoring configurations Access historical data and analytics Control Zabbix through natural language Minimal LLM context usage (3 tools vs 50+ individual endpoints) Installation Quick Install with uvx (Recommended) # Install from GitHub uvx --from git+https://github.com/mpeirone/zabbix-mcp-server@main zabbix-mcp Clone and Install with uv git clone https://github.com/mpeirone/zabbix-mcp-server.git cd zabbix-mcp-server uv sync Docker Installation git clone https://github.com/mpeirone/zabbix-mcp-server.git cd zabbix-mcp-server docker compose up -d Configuration Environment Variables Required: export ZABBIX_URL=https://your-zabbix-server.com Authentication (choose one): # API Token (recommended) export ZABBIX_TOKEN= ${ZABBIX_API_TOKEN} # Or username/password export ZABBIX_USER= ${ZABBIX_USERNAME} export ZABBIX_PASSWORD= ${ZABBIX_PASSWORD} Security: export READ_ONLY= true # Restrict to read operations only export VERIFY_SSL= true # Enable SSL verification export ZABBIX_API_WHITELIST= "host\..*,item\.get,problem\..*" # Allow only specific methods export ZABBIX_API_BLACKLIST= ".*\.delete,.*\.create" # Block specific methods export ZABBIX_API_TIMEOUT=30 # API request timeout in seconds Transport: export ZABBIX_MCP_TRANSPORT=stdio # or streamable-http export ZABBIX_MCP_HOST=127.0.0.1 # HTTP server host export ZABBIX_MCP_PORT=8000 # HTTP server port Debug: export DEBUG= true # Enable verbose logging MCP Client Configuration Claude Desktop (macOS): Edit ~/Library/Application Support/Claude/claude_desktop_config.json : { "mcpServers" : { "zabbix" : { "command" : "uvx" , "args" : [ "--from" , "git+https://github.com/mpeirone/zabbix-mcp-server@main" , "zabbix-mcp" ] , "env" : { "ZABBIX_URL" : "https://zabbix.example.com" , "ZABBIX_TOKEN" : "${ZABBIX_API_TOKEN}" , "READ_ONLY" : "true" } } } } Claude Desktop (Windows): Edit %APPDATA%\Claude\claude_desktop_config.json with the same configuration. Claude Code Integration: claude mcp add zabbix \ -- env ZABBIX_URL=https://your-zabbix-server.com \ -- env ZABBIX_TOKEN= ${ZABBIX_API_TOKEN} \ -- env READ_ONLY= true \ -- uvx --from git+https://github.com/mpeirone/zabbix-mcp-server@main zabbix-mcp The 3 Core Tools 1. zabbix_api - Execute Any Zabbix API Method The main tool for interacting with Zabbix. Supports all 100+ API methods. Signature: zabbix_api(method: str , params: dict = {}) -> dict 2. zabbix_api_docs - Get API Documentation Retrieve documentation for any Zabbix API method. Signature: zabbix_api_docs(method: str ) -> str 3. zabbix_api_list - Discover Available Methods List all available API objects and their methods. Signature: zabbix_api_list( object : str = None ) -> list Common Usage Patterns Querying Hosts Get all hosts with basic info: result = zabbix_api( method= 'host.get' , params={ 'output' : [ 'hostid' , 'host' , 'name' , 'status' ], 'selectInterfaces' : [ 'interfaceid' , 'ip' ] } ) Get hosts in specific group: result = zabbix_api( method= 'host.get' , params={ 'output' : 'extend' , 'groupids' : '2' , 'selectGroups' : 'extend' } ) Search hosts by name: result = zabbix_api( method= 'host.get' , params={ 'output' : [ 'hostid' , 'host' , 'name' ], 'search' : { 'name' : 'web-server' }, 'searchWildcardsEnabled' : True } ) Monitoring Problems and Triggers Get recent problems: result = zabbix_api( method= 'problem.get' , params={ 'output' : 'extend' , 'recent' : True , 'sortfield' : [ 'eventid' ], 'sortorder' : 'DESC' } ) Get active triggers: result = zabbix_api( method= 'trigger.get' , params={ 'output' : 'extend' , 'filter' : { 'value' : 1 }, # Problem state 'skipDependent' : 1 , 'monitored' : True , 'active' : True , 'selectHosts' : [ 'hostid' , 'host' ] } ) Get problems for specific host: result = zabbix_api( method= 'problem.get' , params={ 'output' : 'extend' , 'hostids' : '10084' , 'recent' : True , 'selectAcknowledges' : 'extend' } ) Working with Items and Metrics Get items for a host: result = zabbix_api( method= 'item.get' , params={ 'output' : 'extend' , 'hostids' : '10084' , 'sortfield' : 'name' } ) Get historical data: result = zabbix_api( method= 'history.get' , params={ 'output' : 'extend' , 'history' : 0 , # Float values 'itemids' : '23296' , 'sortfield' : 'clock' , 'sortorder' : 'DESC' , 'limit' : 100 } ) Get latest values: result = zabbix_api( method= 'item.get' , params={ 'output' : [ 'itemid' , 'name' , 'lastvalue' , 'lastclock' ], 'hostids' : '10084' , 'monitored' : True } ) Creating and Managing Resources Create a new host: result = zabbix_api( method= 'host.create' , params={ 'host' : 'server-01.example.com' , 'name' : 'Production Server 01' , 'groups' : [{ 'groupid' : '2' }], 'interfaces' : [{ 'type' : 1 , # Agent 'main' : 1 , 'useip' : 1 , 'ip' : '192.168.1.100' , 'dns' : '' , 'port' : '10050' }] } ) Create a trigger: result = zabbix_api( method= 'trigger.create' , params={ 'description' : 'CPU load too high on {HOST.NAME}' , 'expression' : 'last(/server-01/system.cpu.load[percpu,avg1])>5' , 'priority' : 3 # Average } ) Create an item: result = zabbix_api( method= 'item.create' , params={ 'name' : 'CPU Load' , 'key_' : 'system.cpu.load[percpu,avg1]' , 'hostid' : '10084' , 'type' : 0 , # Zabbix agent 'value_type' : 0 , # Float 'delay' : '60s' } ) Templates and Groups Get templates: result = zabbix_api( method= 'template.get' , params={ 'output' : [ 'templateid' , 'name' ], 'selectGroups' : 'extend' } ) Link template to host: result = zabbix_api( method= 'host.update' , params={ 'hostid' : '10084' , 'templates' : [{ 'templateid' : '10001' }] } ) Get host groups: result = zabbix_api( method= 'hostgroup.get' , params={ 'output' : 'extend' , 'selectHosts' : [ 'hostid' , 'host' ] } ) Discovery and Documentation List all available API objects: result = zabbix_api_list() # Returns: ['action', 'alert', 'apiinfo', 'application', ...] List methods for specific object: result = zabbix_api_list( object = 'host' ) # Returns: ['host.create', 'host.delete', 'host.get', 'host.update', ...] Get documentation for a method: docs = zabbix_api_docs(method= 'host.create' ) # Returns: Detailed documentation with parameters and examples Get API version: result = zabbix_api(method= 'apiinfo.version' , params={}) # Returns: {'result': '7.0.0'} Security Patterns Read-Only Mode For safe exploration and querying without modification risk: export READ_ONLY= true This allows only: *.get , *.version , *.check , *.export methods. Method Whitelisting Allow only specific API methods: # Only allow host and item queries export ZABBIX_API_WHITELIST= "host\.get,item\.get,problem\.get" Method Blacklisting Block dangerous operations: # Block all delete and update operations export ZABBIX_API_BLACKLIST= ".*\.delete,.*\.update,.*\.create" Combined Security export READ_ONLY= true export ZABBIX_API_WHITELIST= "host\..*,item\..*,problem\..*,trigger\..*" export ZABBIX_API_BLACKLIST= "host\.delete,item\.delete" export VERIFY_SSL= true
Agent 识别该技能的关键词,点击任意一个即可复制。

该技能未提供触发词。

下载的 .skill 包内含以下字段。
字段 说明
format格式标识(skill/v1)
skill_id技能唯一 ID
name技能名称
version版本号
description技能描述
category所属分类(数组)
trigger_words触发词列表
tags标签列表
source来源标识
source_url来源链接(本页地址)
exported_at导出时间(每次下载生成)
system_prompt系统提示词正文
model_config模型参数:provider / model / temperature / max_tokens / top_p
examples示例
install_guide各平台导入说明(Coze / Dify / Claude / 自定义框架)
同一份技能可按不同平台格式导出。
.skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用 下载
.skillpro 增强格式,额外含脚本 / 工具 / 依赖 / 钩子占位 下载
.json 纯 JSON 导出,只含 system_prompt 与模型参数 下载
Coze 带 frontmatter 的 Markdown,Coze 平台导入用 下载
Dify Dify DSL,创建应用后直接导入 下载

每日精选 Skill 推荐,免费送到你邮箱

输入邮箱,每天接收一个精选 AI Agent 技能推荐。完全免费,持续更新。

提交后我们会发送一封确认邮件,点击邮件里的链接才会开始收信。

完全免费,取消任意时间。我们不会发送垃圾邮件。