Token导航 LogoToken导航TokenDH.com
Risu Insights logo
运维云端stdio官方级别未说明来源级核验

Risu Insights

MCP Server

一个运行RISU诊断并通过模型上下文协议(MCP)暴露结果的简单服务器,适用于OpenWebUI和其他MCP客户端。

工具数

2

提示词数

0

GitHub Stars

0

资源数

0
Python服务器管理云端部署

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

作者 / 组织

Spectro34

提供方

Spectro34

最后核验

2026/5/17 20:20

运行时

Python

快速接入

先看主来源和安装命令,再打开仓库或文档;下面只保留这个条目的关键接入事实。

命令预览

python3 -m venv .venv

详细介绍

RISU Diagnostics MCP Server

A simple MCP server that runs RISU diagnostics and exposes the results via Model Context Protocol for use with OpenWebUI and other MCP clients.

How It Works

The server executes RISU through Ansible for all hosts (including localhost):

For all hosts (localhost and remote):

  • Uses Ansible command (ansible -m shell) to connect and execute RISU # to-do use risu ansible module
  • Leverages Ansible's connection system (local, SSH, paramiko, etc.) automatically
  • For localhost, uses ansible_connection=local from inventory
  • Retrieves the JSON output from Ansible's command output
  • Parses and returns the results

Inventory:

  • Uses Ansible-style inventory files (INI format) to resolve hostnames
  • Reads Ansible-style variables (ansible_user, ansible_port, ansible_become, ansible_connection, etc.) from inventory
  • Passes these variables to Ansible command, which handles all connection details

The MCP server exposes two tools:

  • show_inventory - Reads and displays the Ansible-style inventory file
  • run_diagnostics - Executes RISU, parses results, and returns formatted reports

Installation

cd risu-insights
python3 -m venv .venv
source .venv/bin/activate
pip install -e .

Ensure risu is installed on the managed nodes

zypper addrepo https://download.opensuse.org/repositories/home:hsharma/openSUSE_Factory/home:hsharma.repo
zypper refresh
zypper install risu

Inventory Path:

  • Use --inventory /path/to/inventory to specify a custom inventory file when starting the server
  • Default inventory path: ./inventory/hosts (relative to project root) if not specified
  • You can also override the inventory path per tool call using the inventory parameter

Inventory Path Resolution (priority order):

  1. Path specified in tool call (inventory parameter)
  2. --inventory CLI option when starting server
  3. RISU_DIAG_INVENTORY environment variable
  4. Default ./inventory/hosts (relative to project root)

Adding to OpenWebUI

  1. Start the RISU diagnostics server with mcpo bridge:
   pip install mcpo
   mcpo --port 8000 --host 0.0.0.0 -- risu-insights-stdio
   
   # Optional mention custom inventory path
    --inventory /path/to/inventory
  1. In OpenWebUI, go to Settings → Tools → Tool Servers
  1. Click Add Server and configure:

- Type: OpenAPI - URL: http://localhost:8000 - OpenAPI Spec: openapi.json - Auth: None

  1. Click Verify to test the connection
  1. Save the server configuration
  1. In any chat, enable the tool and use:

- show_inventory - View available hosts - run_diagnostics - Run RISU diagnostics on specified hosts with optional plugin filters

Usage Examples

  • Run diagnostics on localhost: run_diagnostics(hosts="localhost")
  • Run with plugin filter: run_diagnostics(hosts="localhost", plugin_filter="core")
  • Run on multiple hosts: run_diagnostics(hosts="host1,host2")
  • Run on all hosts: run_diagnostics(hosts="all")
  • Use custom inventory: run_diagnostics(hosts="localhost", inventory="/path/to/custom/inventory")
  • Show inventory from custom path: show_inventory(inventory="/path/to/custom/inventory")

目录标签

目录标签

Python服务器管理云端部署诊断工具本地部署Ansible集成RISU诊断MCP协议

接入字段

传输方式(transport,传输协议)

stdio

鉴权方式(authType,认证方式)

none

运行时(runtime,运行环境)

Python

工具数量(toolCount,工具数)

2

资源数量(resourceCount,资源数)

0

提示词数量(promptCount,提示词数)

0

权限和风险

stdionone部署方式未说明

接入前请确认传输方式、认证方式和部署位置,并根据实际工具能力限制访问范围。

安装前确认

不要直接授予不必要的文件、网络或账号权限;先核对安装命令和配置内容。

来源信息

继续浏览同类 MCP