Token导航 LogoToken导航

Agent Skill · 研究检索

modelwise-openclaw-setup(modelwise OpenClaw 设置)

modelwise-openclaw-setup 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

modelwise-openclaw-setup 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为接触敏感数据;许可证为 MIT-0,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 15,565。

总安装 7,760周安装 317GitHub Stars 公开资料未说明下载 2,485核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
1.0.1

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:modelwise-openclaw-setup(modelwise OpenClaw 设置)
来源仓库:https://github.com/deerleo/modelwise-openclaw-setup
安装命令:
openclaw skills install modelwise-openclaw-setup
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install modelwise-openclaw-setup

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于响应用户请求,指导安装、配置和启动 OpenClaw 网关及相关服务。适合新用户入门或现有环境重建时的快速部署场景。modelwise-openclaw-setup 属于研究检索类 Skill,可作为该场景下的辅助能力补充。通过 clawhub 安装,需确认系统环境和依赖项。

建议结合原始 README 核验安装脚本和安全检查步骤。使用前请评估是否会触发命令执行或网络下载,避免安全风险。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install modelwise-openclaw-setup;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要接触敏感数据,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

modelwise-openclaw-setup 怎么安装?

在终端执行 openclaw skills install modelwise-openclaw-setup。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

modelwise-openclaw-setup 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

modelwise-openclaw-setup 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

modelwise-openclaw-setup 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

modelwise-openclaw-setup 用的人多吗?

公开数据是累计安装 7,760 次,近一周安装 317 次,下载 2,485 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

modelwise-openclaw-setup 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
modelwise-openclaw-setup

modelwise OpenClaw 设置 · 本页

7,760—接触敏感数据MIT-0通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
OpenClaw Setup
description
This skill should be used when the user asks to "install OpenClaw", "setup OpenClaw", "configure OpenClaw", "start OpenClaw gateway", "部署 OpenClaw", or needs help with OpenClaw installation, configuration, and initial setup on Windows/macOS/Linux systems. By ModelWise team.
version
1.0.0

OpenClaw Installation and Setup Guide

By ModelWise team - Professional AI Agent Solutions

This skill provides comprehensive guidance for installing, configuring, and running OpenClaw - an open-source personal AI assistant gateway.

Overview

OpenClaw is a multi-channel AI assistant gateway that enables users to use AI assistants on their existing messaging platforms (WhatsApp, Telegram, Slack, Discord, etc.).

Key Features:

  • Local-first architecture
  • 20+ messaging channel integrations
  • Multi-model support (Anthropic, OpenAI, Google, local models)
  • Voice interaction (Voice Wake + Talk Mode)
  • Canvas visualization workspace (A2UI protocol)
  • Skills extension system

Prerequisites

System Requirements

PlatformRequirements
WindowsWindows 10/11, Node.js >= 22.12.0
macOSmacOS 12+, Node.js >= 22.12.0
LinuxUbuntu 20.04+/Debian 11+, Node.js >= 22.12.0

Node.js Installation

Check Node.js version first:

node --version
Windows
# Option 1: Using winget
winget install OpenJS.NodeJS.LTS

# Option 2: Using Chocolatey
choco install nodejs-lts

# Option 3: Download installer
# https://nodejs.org/
macOS / Linux
# Install nvm
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.1/install.sh | bash

# Reload shell
source ~/.zshrc  # or ~/.bashrc

# Install Node.js 22
nvm install 22
nvm use 22
nvm alias default 22

Installation

Step 1: Install OpenClaw CLI

# All platforms
npm install -g openclaw@latest

# Verify installation
openclaw --version

Step 2: Run Configuration Wizard

# Start interactive configuration
openclaw onboard

# Or with daemon installation (recommended)
openclaw onboard --install-daemon

The wizard will guide through:

  1. Model provider selection (Anthropic, OpenAI, Volcengine, etc.)
  2. API key configuration
  3. Channel setup (Telegram, Slack, etc.)
  4. Gateway configuration

Step 3: Configure API Keys

API keys are stored in ~/.openclaw/credentials/ (or %USERPROFILE%\.openclaw\credentials\ on Windows).

# Set Anthropic API key
openclaw credentials set anthropic YOUR_API_KEY

# Set Volcengine API key (for Kimi)
openclaw credentials set volcengine YOUR_API_KEY

# Set OpenAI API key
openclaw credentials set openai YOUR_API_KEY

Step 4: Start Gateway

# Start gateway on default port (18789)
openclaw gateway

# Or specify port
openclaw gateway --port 18789

# Run in background (macOS/Linux)
openclaw gateway --daemon

# Windows: run in new window or use pm2
start /B openclaw gateway

Configuration

Config File Location

PlatformPath
macOS/Linux~/.openclaw/openclaw.json
Windows%USERPROFILE%\.openclaw\openclaw.json

Essential Configuration

{
  // Gateway settings
  gateway: {
    enabled: true,
    port: 18789,
    mode: "local",
    bind: "0.0.0.0"
  },

  // Agent settings
  agents: {
    defaults: {
      model: "volcengine/kimi-k2.5",
      thinking: {
        enabled: true,
        level: "medium"
      }
    }
  },

  // Telegram channel (example)
  telegram: {
    enabled: true,
    botToken: "YOUR_BOT_TOKEN",
    dmPolicy: "open",
    streaming: "partial"
  }
}

Model Provider Configuration

ProviderModel FormatCredentials Key
Anthropicanthropic/claude-opus-4-6anthropic
OpenAIopenai/gpt-5.2openai
Volcenginevolcengine/kimi-k2.5volcengine
Googlegoogle/gemini-2.5-progoogle
Locallocal/llama-3.2N/A

Channel Setup

Telegram Bot Setup

  1. Create Bot:

- Open @BotFather on Telegram - Send /newbot - Follow instructions to get bot token

  1. Configure OpenClaw:
openclaw credentials set telegram 7123456789:AAHxxxxxxxx
openclaw config set telegram.enabled true
openclaw config set telegram.dmPolicy "open"
  1. Restart Gateway:
openclaw gateway restart

Security Policies

DM Policy Options:

  • open: Accept messages from anyone (for public bots)
  • pairing: Require pairing code approval (default, more secure)

Common Commands

Diagnostics

# Check system status
openclaw doctor

# View running processes (macOS/Linux)
ps aux | grep openclaw | grep -v grep

# View running processes (Windows PowerShell)
Get-Process | Where-Object {$_.ProcessName -eq "node"}

# Check gateway status
openclaw gateway status

Updates

# Update to latest stable
openclaw update --channel stable

# Update to beta version
openclaw update --channel beta

Maintenance

# Stop gateway (macOS/Linux)
killall openclaw openclaw-gateway

# Stop gateway (Windows PowerShell)
Stop-Process -Name "node" -Force

# View logs
tail -f ~/.openclaw/logs/gateway.log  # macOS/Linux
Get-Content $env:USERPROFILE\.openclaw\logs\gateway.log -Tail 100  # Windows

# Clear sessions
openclaw sessions prune

Directory Structure

PlatformBase Path
macOS/Linux~/.openclaw/
Windows%USERPROFILE%\.openclaw\
.openclaw/
├── openclaw.json       # Main configuration
├── credentials/        # API keys (encrypted)
├── sessions/           # Session data
├── logs/               # Log files
├── workspace/          # Working directory
│   └── skills/         # Custom skills
├── agents/             # Agent configurations
├── canvas/             # Canvas data
├── browser/            # Browser profiles
└── memory/             # Memory storage

Troubleshooting

Common Issues

1. Node.js version mismatch:

# Check version
node --version  # Should be >= 22.12.0

# Fix (macOS/Linux)
nvm use 22

# Fix (Windows)
# Download and install from https://nodejs.org/

2. Port already in use:

# macOS/Linux
lsof -i :18789
kill -9 <PID>

# Windows PowerShell
netstat -ano | findstr :18789
taskkill /PID <PID> /F

3. Permission denied:

# macOS/Linux
sudo chown -R $(whoami) ~/.npm

# Windows (run PowerShell as Administrator)
npm cache clean --force

4. Gateway not starting:

# Check logs (macOS/Linux)
cat ~/.openclaw/logs/gateway.log

# Check logs (Windows)
type %USERPROFILE%\.openclaw\logs\gateway.log

# Run diagnostics
openclaw doctor

Reset Configuration

# Backup and reset (macOS/Linux)
cp ~/.openclaw/openclaw.json ~/.openclaw/openclaw.json.bak
rm -rf ~/.openclaw
openclaw onboard

# Backup and reset (Windows PowerShell)
Copy-Item $env:USERPROFILE\.openclaw\openclaw.json $env:USERPROFILE\.openclaw\openclaw.json.bak
Remove-Item -Recurse -Force $env:USERPROFILE\.openclaw
openclaw onboard

Quick Reference

# Installation
npm install -g openclaw@latest

# Configuration
openclaw onboard --install-daemon

# Start/Stop
openclaw gateway              # Start
killall openclaw              # Stop (macOS/Linux)

# Status
openclaw doctor               # Diagnostics
openclaw --version            # Version

# Update
openclaw update --channel stable

Additional Resources

Scripts

Cross-platform installation checkers:

  • scripts/check-installation.sh - Bash script (macOS/Linux/Windows Git Bash)
  • scripts/check-installation.ps1 - PowerShell script (Windows)

Reference Files

  • references/configuration-reference.md - Complete configuration options

Example Files

  • examples/openclaw.json - Full configuration example
  • examples/telegram-setup.md - Telegram channel setup guide

Official Resources

  • Website: https://openclaw.ai
  • Documentation: https://docs.openclaw.ai
  • GitHub: https://github.com/openclaw/openclaw
  • Discord: https://discord.gg/clawd
  • ClawHub (Skills): https://clawhub.com

平台分布

OpenClaw

88.94%
按下载量换算2,210

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills