ZenOps
🧘 通过钉钉、飞书、企微智能机器人用自然语言查询运维资源的工具。
claude mcp add --transport stdio opsre-zenops docker run -itd --name zenops -p 8080:8080 -p 8081:8081 -v ./data:/app/data docker.cnb.cool/opsre/zenops
How to use
ZenOps is a data-driven operational query tool that exposes multiple access methods for querying cloud resources, CI/CD tooling, and other operational data. It supports a unified MCP protocol proxy, a RESTful HTTP API, a command-line interface (CLI) built on Cobra, and a web frontend. Through MCP, you can configure external MCP clients to route queries to ZenOps and receive structured responses. The project also provides integrations with chat tools (DingTalk, Feishu, and WeChat) for conversational, streaming query results, enabling operators to ask questions in natural language and receive live updates.
To use ZenOps, deploy it via Docker as shown in the quick-start guide. Once running, open the frontend at http://localhost:8080 to configure cloud vendors, CI/CD connectors (e.g., Jenkins), and LLM/IM credentials. You can use the HTTP API to programmatically query resources, the CLI for local scripted queries, or the MCP interface to connect external MCP clients. The MCP support allows external systems to proxy queries through ZenOps, enabling centralized access to multi-cloud resources and automation tooling within your environment.
How to install
Prerequisites:
- Docker installed and running on your host
- Access to the internet to pull the ZenOps image
Step-by-step installation:
-
Install Docker (if not already installed):
- macOS: https://docs.docker.com/desktop/mac/
- Windows: https://docs.docker.com/desktop/windows/
- Linux: follow your distro's instructions (e.g., apt/yum install docker)
-
Run ZenOps with Docker (quick-start):
# Deploy ZenOps in detached mode with ports and data volume
docker run -itd \
--name zenops \
-p 8080:8080 \
-p 8081:8081 \
-v ./data:/app/data \
docker.cnb.cool/opsre/zenops
- Verify startup:
docker ps | grep zenops
- Access the UI/API:
- Frontend: http://localhost:8080
- Admin default credentials (per README): admin/admin123
Additional notes
Tips and common issues:
- The MCP proxy is enabled by default after the container starts; you can configure external MCP clients to route queries to ZenOps.
- If you change configuration in the UI, ensure the ZenOps container has access to the updated credentials/keys and that network access to required cloud APIs is allowed.
- When upgrading, pull the latest image and recreate the container while preserving the ./data volume to keep configurations and state.
- If you run into port conflicts, adjust the -p mappings in the docker run command to free up required ports.
- Environment variables are commonly used to inject API keys or toggles; this container reads configuration from its mounted volume and its web UI settings.
Related MCP Servers
NagaAgent
A simple yet powerful agent framework for personal assistants, designed to enable intelligent interaction, multi-agent collaboration, and seamless tool integration.
SearChat
Search + Chat = SearChat(AI Chat with Search), Support OpenAI/Anthropic/VertexAI/Gemini, DeepResearch, SearXNG, Docker. AI对话式搜索引擎,支持DeepResearch, 支持OpenAI/Anthropic/VertexAI/Gemini接口、聚合搜索引擎SearXNG,支持Docker一键部署。
Archive-Agent
Find your files with natural language and ask questions.
winremote
Windows Remote MCP Server — 40+ tools for desktop automation, process management, file operations via FastMCP
jenkins -enterprise
The most advanced Jenkins MCP server available - Enterprise debugging, multi-instance management, AI-powered failure analysis, vector search, and configurable diagnostics for complex CI/CD pipelines.
mcp-dingding-bot
MCP Server for send text/markdown message via dingding (aka dingtalk) group custom robot