Glances

v0.1.0

Glances - 跨平台系统监控工具

0· 80·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for cn-big-cabbage/glances.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Glances" (cn-big-cabbage/glances) from ClawHub.
Skill page: https://clawhub.ai/cn-big-cabbage/glances
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install glances

ClawHub CLI

Package manager switcher

npx clawhub@latest install glances
Security Scan
Capability signals
Crypto
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
Purpose & Capability
Name/description (Glances system monitor) match the provided content. All commands, config examples, and dependencies mentioned (pip, bottle, docker, InfluxDB, Prometheus, systemd, etc.) are expected for installing and operating Glances.
Instruction Scope
SKILL.md contains step‑by‑step installation and runtime instructions and explicitly states the AI may 'execute installation commands' and 'start Glances' — which is consistent with an installer helper but grants the agent discretion to run shell commands. The guide includes operations that read system state and write files (create config files, write service units, create scripts, add cron jobs, inspect sockets like /var/run/docker.sock) — all relevant to monitoring but worth reviewing because they affect system state and may require root privileges.
Install Mechanism
This is instruction-only (no install spec, no downloaded archives). The instructions rely on standard package managers (pip, apt, brew, docker) and the official Docker image; nothing in the files points to unusual or remote installers.
Credentials
The skill does not request environment variables or external credentials up front, which is proportionate. However the docs include examples that store credentials in plaintext config (SMTP password, InfluxDB token: 'mytoken', etc.) and suggest enabling network‑facing Web/API modes. The skill does not declare or manage secrets — if you use these features you'll need to provide credentials and should avoid plaintext or unsafe storage.
Persistence & Privilege
The guides show creating systemd units, placing scripts under /usr/local/bin, adding cron jobs, and running containers with host mounts (e.g., /var/run/docker.sock). These are normal for a monitoring agent but are privileged operations that modify system configuration and can increase blast radius; they require explicit user consent and careful configuration (user account, network binding, auth).
Assessment
This skill is a coherent, instruction-only helper for Glances, but its advice includes actions that change system state and may expose services or credentials. Before you let an AI run these steps or apply them yourself: 1) Review any commands the agent proposes to run (especially sudo/systemd/cron file writes). 2) Prefer isolated installs (pipx or Docker) if you want less host impact. 3) Do not bind the Web/API to 0.0.0.0 without authentication and firewall rules — enable username/password or restrict access via firewall/SSH tunnel. 4) Avoid storing passwords/tokens in plaintext config; use dedicated secrets management or environment variables with restricted file permissions. 5) If monitoring Docker, note that mounting /var/run/docker.sock grants broad control over containers. 6) If you want higher assurance, run installation steps manually or in an isolated environment and inspect generated config/service files before enabling them.

Like a lobster shell, security has layers — review code before you run it.

Runtime requirements

📊 Clawdis
latestvk97cts8kanakse8kmwcchpt0a984wwa8
80downloads
0stars
1versions
Updated 1w ago
v0.1.0
MIT-0

Glances 跨平台系统监控工具

技能概述

本技能帮助用户使用 Glances 进行系统性能监控与分析,支持以下场景:

  • 实时监控: 在终端实时查看 CPU、内存、磁盘、网络等资源使用情况
  • Web 界面: 通过浏览器远程访问系统监控仪表板
  • API 服务: 通过 REST API 获取系统指标数据,便于集成
  • 告警配置: 设置资源阈值,当超过时自动触发告警
  • 导出分析: 将监控数据导出到 CSV、InfluxDB、Prometheus 等存储系统
  • 服务器模式: 以客户端-服务端模式监控远程主机

支持平台: Linux、macOS、Windows、FreeBSD,支持 Docker 容器化部署

使用流程

AI 助手将引导你完成以下步骤:

  1. 安装 Glances(如未安装)
  2. 以合适的模式启动监控(终端/Web/API)
  3. 解读监控数据和告警信息
  4. 根据需要配置自定义参数
  5. 集成到现有工作流或监控系统

关键章节导航

AI 助手能力

当你向 AI 描述监控需求时,AI 会:

  • 自动检测系统环境并选择合适的安装方式
  • 启动 Glances 并解读当前系统状态
  • 识别高负载进程和资源瓶颈
  • 配置 Web 服务器模式并提供访问地址
  • 设置导出插件将指标写入时序数据库
  • 分析告警日志并给出优化建议
  • 生成监控配置文件以满足特定需求

核心功能

  • 实时监控 CPU、内存、磁盘 I/O、网络流量
  • 进程列表(支持排序、过滤、终止)
  • Web UI 仪表板(内置 HTTP 服务器)
  • REST API / XML-RPC API 远程数据获取
  • 客户端-服务端模式(通过 SSH 隧道监控远程主机)
  • 告警系统(Careful / Warning / Critical 三级)
  • 插件化架构(支持 30+ 内置插件)
  • 导出支持:CSV、InfluxDB、Prometheus、Elasticsearch、MQTT 等
  • Docker 和 Kubernetes 容器监控
  • 支持 Python API 二次开发

快速示例

# 启动终端交互模式
glances

# 启动 Web 服务器模式(浏览器访问)
glances -w

# 以服务端模式运行(等待客户端连接)
glances -s

# 连接远程服务端
glances -c 192.168.1.100

# 以 REST API 模式运行
glances -w --disable-webui

# 查看特定进程
glances --process-filter python

# 将数据导出到 CSV
glances --export csv --export-csv-file /tmp/glances.csv

# 静默模式(仅记录日志,无界面)
glances -q --export influxdb

监控指标说明

指标类别描述告警颜色
CPU 使用率总体和每核心使用率绿/蓝/黄/红
内存 / Swap物理内存和交换空间使用绿/蓝/黄/红
磁盘 I/O读写速率和等待时间绿/蓝/黄/红
网络接口上下行速率和数据包数绿/蓝/黄/红
进程列表CPU/内存占用最高的进程可排序
系统负载1/5/15 分钟平均负载绿/蓝/黄/红
传感器CPU 温度(需要硬件支持)绿/黄/红
Docker容器状态和资源使用绿/红

告警级别

Glances 使用四种颜色标识资源状态:

  • 绿色 (OK): 正常,无需关注
  • 蓝色 (Careful): 需要关注,资源使用偏高
  • 黄色 (Warning): 警告,资源使用较高
  • 红色 (Critical): 严重,需要立即处理

默认阈值可通过配置文件 ~/.config/glances/glances.conf 自定义。

安装要求

  • Python 3.8 或更高版本
  • pip 包管理器
  • (可选)psutil >= 5.3.0(核心依赖,通常自动安装)
  • (可选)bottles/docker SDK(容器监控)
  • (可选)InfluxDB / Prometheus 客户端(数据导出)

项目链接

Comments

Loading comments...