Win10 安装OpenClaw 龙虾

Win10 安装OpenClaw 龙虾

安装方法

前置环境安装

安装OpenClaw

  • 以管理身份运行CMD控制台
SHELL 复制代码
npm install -g openclaw@latest

配置OpenClaw

  • 打开配置仪表盘
SHELL 复制代码
openclaw onboard --install-daemon

OpenClaw仪表盘详细配置

SHELL 复制代码
# 是否修复配置文件
Run "openclaw doctor --fix" now? [Y/n] Y
# 我明白这是默认的个人使用,共享/多用户使用需要锁定。要继续吗?
I understand this is personal-by-default and shared/multi-user use requires lock-down. Continue?
|  > Yes
|    No
# 是否帮助改进,上传日常使用信息
Help make OpenClaw better?
|  > No thanks
|    Yes, share feature stats
# 设置模式
Setup mode
|    Keep existing model config
|  > QuickStart (recommended) (Recommended local setup. Change details later with openclaw configure.)
|    Manual setup
|    Import from another agent
# 模型/身份验证提供者(根据自己情况选择,我使用的是本地Ollama)
 Model/auth provider
|    OpenAI
|    OpenRouter
|    xAI (Grok)
|    Google
|    Anthropic
|  > More...
|    Skip for now
# 模型/身份验证提供者(根据自己情况选择,我使用的是本地Ollama)
Model/auth provider
|
|  Search:
|  ...
|    DeepInfra
|    DeepSeek
|    Featherless AI
|    Fireworks
|    GMI Cloud
|    Google Vertex
|    Groq
|    Hugging Face
|    Kilo Gateway
|    LiteLLM
|    LM Studio
|    LongCat
|    Meta
|    Microsoft Foundry
|    MiniMax
|    Mistral AI
|    Moonshot AI (Kimi K2.6)
|    NovitaAI
|    NVIDIA
|  > Ollama (currently configured) (Cloud and local open models)
|    OpenCode
|  ...
# Ollama 认证方式
Ollama auth method
|    Keep current config
|  > Ollama (Connect to an Ollama server and select a cloud or local model)
|    Ollama Cloud
|    Back
# Ollama 模式
Ollama mode
|    Cloud + Local
|  > Local only (Local models only)
# Ollama base URL
Ollama base URL
|  http://127.0.0.1:11434
# 是否重试Ollama服务器地址?
Retry this Ollama address now?
|  > Yes /   No
# 默认模式
Default model
|    Keep current (ollama/qwen3-coder)
|  > Enter model manually
|    Browse all models
# 默认模式(根据自己情况选择)
Default model
|  ollama/qwen3-coder
# 现在通过实时完成测试AI访问?
Test AI access now with a live completion?
|  > Yes /   No
# 您想如何继续?
How would you like to continue?
|    Fix model/auth setup
|  > Continue anyway
# 选择频道(快速启动)
Select channel (QuickStart)
|
|  Search: _
|  > Skip for now (You can add channels later via `openclaw channels add`)
|    A2A (Agent-to-Agent Protocol)
|    AgentMail (Email)
|    Buzz
|    ClickClack
|    Discord (Bot API)
|    Feishu/Lark (飞书)
|    Google Chat (Chat API)
|    iMessage (imsg)
|    IRC (Server + Nick)
|    LINE (Messaging API)
|    Matrix (plugin)
|    Mattermost (plugin)
|    Microsoft Teams (Teams SDK)
|    Nextcloud Talk (self-hosted)
|    Nostr (NIP-04 DMs)
|    QQ Bot (Official API)
|    Raft (CLI wake bridge)
|    Reef
|    Signal (signal-cli)
|    Slack (Socket Mode)
|    SMS (Twilio)
|    Synology Chat (Webhook)
|  ...
# 搜索提供程序
Search provider
|
|  Search: _
|    Brave Search (Brave Search web results. · Brave Search API key required)
|    DuckDuckGo Search (experimental) (Free web search fallback with no API key required · key-free)
|    Exa Search (Neural + keyword search with date filters and content extraction · Exa API key required)
|    Firecrawl Search (Structured results with optional result scraping · Firecrawl API key required)
|  > Firecrawl Search (Free) (Free web search via Firecrawl's hosted starter tier --- no API key required · key-free)
|    Gemini (Google Search) (Requires Google Gemini API key · Google Search grounding · Google Gemini API key required)
|    Grok (xAI) (Uses xAI OAuth or API key · xAI web-grounded responses · xAI API key required)
|    Kimi (Moonshot) (Requires Moonshot / Kimi API key · Moonshot web search · Moonshot / Kimi API key required)
|    MiniMax Search (Structured results via MiniMax Token Plan search API · MiniMax Token Plan key or OAuth token
|  required)
|    Ollama Web Search (Local Ollama host · requires ollama signin · key-free)
|    Parallel Search (LLM-optimized dense excerpts from web sources · Parallel API key required)
|    Parallel Search (Free) (Free web search via Parallel's hosted Search MCP --- no API key required · key-free)
|    Perplexity Search (Requires Perplexity API key or OpenRouter API key · structured results · Perplexity API key
|  required)
|    SearXNG Search (Self-hosted meta-search with no API key required · SearXNG Base URL required)
|    Tavily Search (Structured results with domain filters and AI answer summaries · Tavily API key required)
|    Skip for now
|  ↑/↓ to select • Enter: confirm • Type: to search
# 安装Firecrawl插件?
Install Firecrawl plugin?
|  > Download from npm (@openclaw/firecrawl-plugin)
|    Download from ClawHub (clawhub:@openclaw/firecrawl-plugin)
|    Skip for now
# 网关服务已安装
Gateway service already installed
|    Restart
|    Reinstall
|  > Skip

测试

相关推荐
IDIOT___IDIOT6 小时前
DSH 对话回收站(dsh-session-trash)
ai·dsh
玩AI的奶茶6 小时前
MiniMax H3 出 30 秒长片:显存账、拼接法和一次的成本
大数据·人工智能·ai·gpu算力·算力租赁
犀思云6 小时前
从SD-WAN到AI WAN:企业广域网能力演进与架构变化解读
ai·sd-wan·wan·企业广域网
犀思云8 小时前
AI WAN实践:分布式AI训练与推理如何通过智能广域网提升效率?
ai·wan·分布式ai训练·分布式ai推理·智能广域网
Maynor9968 小时前
ChatGPT Pro 额度档位怎么理解:5x / 20x / 50x 使用笔记
ai·chatgpt·教程
吴名氏.9 小时前
Java转AI Agent教程(第一周)让模型先跑起来
人工智能·python·ai·ai agent
FPGA小迷弟9 小时前
FPGA 与无人机:从飞控实时性到智能载荷的硬件加速实践
人工智能·ai·fpga开发·架构·无人机·ai编程
zhangphil10 小时前
Agent决定做什么,Skill定义怎么做,MCP让它能调用外部能力
ai
玫瑰互动GEO11 小时前
GEM优化广告六要素: 从命名一致到落地页双对齐的工程化落地
人工智能·ai·chatgpt·gem优化