【养虾之路】OpenClaw 完全新手指南(麒麟Ky10SP2安装)

本文详细记录了在国产麒麟KY10服务器操作系统上部署OpenClaw并行计算框架的全过程。作为面向高性能计算场景的分布式系统，OpenClaw的安装涉及底层硬件驱动适配、依赖库编译优化及集群网络配置等关键技术环节。麒麟KY10作为基于Linux的国产服务器平台，其安全增强内核与OpenClaw的异构计算架构存在特定的兼容性要求，需通过定制化编译参数解决指令集优化问题。实施过程严格遵循从基础环境准备

weixin_50902636

1207人浏览 · 2026-03-10 10:52:02

weixin_50902636 · 2026-03-10 10:52:02 发布

文章目录

OpenClaw 完全新手指南

OpenClaw 完全新手指南

个人 AI 助手本地部署与配置权威指南

EXFOLIATE! EXFOLIATE!

🚀什么是 OpenClaw？

OpenClaw 是一款个人 AI 助手，运行在你自己的设备上。它可以通过你已经在使用的通讯渠道（WhatsApp、Telegram、Slack、Discord、Google Chat、Signal、iMessage、Microsoft Teams、WebChat 等）与你交互，还支持语音对话、Canvas 可视化等高级功能。

OpenClaw 是一个本地优先的 AI 助手网关，它的核心设计理念是：

个人化：为你个人服务，而非企业级多用户系统
本地运行：Gateway 运行在你的设备上（macOS、Linux、Windows WSL2）
多渠道接入：支持几乎所有主流通讯平台
功能丰富：浏览器控制、Canvas 可视化、语音交互、定时任务等

架构概览

WhatsApp / Telegram / Slack / Discord / Google Chat / Signal / iMessage / WebChat
               │
               ▼
┌───────────────────────────────┐
│            Gateway            │
│       (控制平面 - 核心)        │
│     ws://127.0.0.1:18789      │
└──────────────┬────────────────┘
               │
               ├─ AI 智能体 (Agent)
               ├─ CLI 命令行工具
               ├─ WebChat 网页界面
               ├─ macOS/iOS/Android 客户端
               └─ 浏览器控制 (Chrome CDP)

📦安装指南

系统要求

Node.js ≥ 22（必须）
操作系统：macOS、Linux、Windows (WSL2)
包管理器：npm、pnpm 或 bun
依赖：cmake(>=3.19)、git、gcc(高版本)
当前服务器配置：8C8G100G

快速安装（推荐）

macOS / Linux

curl -fsSL https://openclaw.ai/install.sh | bash

Windows (PowerShell)

iwr -useb https://openclaw.ai/install.ps1 | iex

npm 安装

# 使用 npm
npm install -g openclaw@latest

# 或使用 pnpm
pnpm add -g openclaw@latest

麒麟ky10_x86_64操作系统安装openclaw

检查系统版本

[root@k8s-master ~]# uname -a
Linux k8s-master.deploy.local 4.19.90-24.4.v2101.ky10.x86_64 #1 SMP Mon May 24 12:14:55 CST 2021 x86_64 x86_64 x86_64 GNU/Linux

安装conda

# 下载 Miniconda 安装脚本（适用于 Linux x86_64）
[root@k8s-master build-linux-x64]# wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
[root@k8s-master build-linux-x64]# bash Miniconda3-latest-Linux-x86_64.sh
[root@k8s-master build-linux-x64]# source ~/.bashrc
[root@k8s-master build-linux-x64]# conda create -n openclaw-build python=3.10
[root@k8s-master build-linux-x64]# conda activate openclaw-build
(openclaw-build) [root@k8s-master build-linux-x64]#conda install -c conda-forge gcc=10.4.0 gxx=10.4.0 make cmake git nodejs=24
# 将 Conda 环境的 bin 目录加入 PATH（激活环境后通常已自动处理，但可手动确认）
(openclaw-build) [root@k8s-master build-linux-x64]# export PATH=$CONDA_PREFIX/bin:$PATH

# 设置库路径，让运行时使用 Conda 的库
(openclaw-build) [root@k8s-master build-linux-x64]# export LD_LIBRARY_PATH=$CONDA_PREFIX/lib:$LD_LIBRARY_PATH

# 显式告诉 CMake 使用 Conda 的 make
(openclaw-build) [root@k8s-master build-linux-x64]# export CMAKE_MAKE_PROGRAM=$CONDA_PREFIX/bin/make

# 可选：指定编译器（确保 CMake 使用 Conda 的 gcc/g++）
(openclaw-build) [root@k8s-master build-linux-x64]# export CC=$CONDA_PREFIX/bin/cc
(openclaw-build) [root@k8s-master build-linux-x64]# export CXX=$CONDA_PREFIX/bin/c++

# 应显示 ~/miniconda3/envs/openclaw-build/bin/make
(openclaw-build) [root@k8s-master build-linux-x64]# which make
/root/miniconda3/envs/openclaw-build/bin/make

# 应显示 ~/miniconda3/envs/openclaw-build/bin/cmake
(openclaw-build) [root@k8s-master build-linux-x64]# which cmake
/root/miniconda3/envs/openclaw-build/bin/cmake
# 应显示 ~/miniconda3/envs/openclaw-build/bin/x86_64-conda-linux-gnu-cc（或类似
(openclaw-build) [root@k8s-master build-linux-x64]# which gcc
/root/miniconda3/envs/openclaw-build/bin/gcc

(openclaw-build) [root@k8s-master build-linux-x64]# echo $CMAKE_MAKE_PROGRAM
/root/miniconda3/bin/make

(openclaw-build) [root@k8s-master build-linux-x64]# npm cache clean --force
(openclaw-build) [root@k8s-master build-linux-x64]# npm uninstall -g openclaw

进入 `node-llama-cpp` 目录

(openclaw-build) [root@k8s-master build-linux-x64]# cd /root/miniconda3/envs/openclaw-build/lib/node_modules/openclaw/node_modules/node-llama-cpp

# 创建 llama 目录并克隆 llama.cpp
(openclaw-build) [root@k8s-master node-llama-cpp]# mkdir -p llama
(openclaw-build) [root@k8s-master node-llama-cpp]# cd llama
(openclaw-build) [root@k8s-master llama]# git clone --depth 1 https://github.com/ggml-org/llama.cpp.git
(openclaw-build) [root@k8s-master llama]# cd llama.cpp
(openclaw-build) [root@k8s-master llama]# git tag | grep b8121

设置必要的环境变量（确保使用Conda的工具）

(openclaw-build) [root@k8s-master build-linux-x64]# export CC=/root/miniconda3/envs/openclaw-build/bin/cc
(openclaw-build) [root@k8s-master build-linux-x64]# export CXX=/root/miniconda3/envs/openclaw-build/bin/c++
(openclaw-build) [root@k8s-master build-linux-x64]# export CMAKE_MAKE_PROGRAM=/root/miniconda3/envs/openclaw-build/bin/make
(openclaw-build) [root@k8s-master build-linux-x64]# export LD_LIBRARY_PATH=/root/miniconda3/envs/openclaw-build/lib:$LD_LIBRARY_PATH

创建构建目录并进入

(openclaw-build) [root@k8s-master build-linux-x64]# mkdir -p build-linux-x64 && cd build-linux-x64

使用 Conda 的 cmake 配置项目（关键步骤）

执行以下命令，显式指定编译器、make 程序，并传递所有必要的编译选项

(openclaw-build) [root@k8s-master build-linux-x64]# sjcmake \
  -DCMAKE_C_COMPILER=/root/miniconda3/envs/openclaw-build/bin/cc \
  -DCMAKE_CXX_COMPILER=/root/miniconda3/envs/openclaw-build/bin/c++ \
  -DCMAKE_MAKE_PROGRAM=/root/miniconda3/envs/openclaw-build/bin/make \
  -DCMAKE_BUILD_TYPE=Release \
  -DGGML_BUILD_NUMBER=1 \
  -DNLC_CURRENT_PLATFORM=linux-x64 \
  -DNLC_TARGET_PLATFORM=linux-x64 \
  -DNLC_VARIANT=b8121 \
  -DGGML_METAL=OFF \
  -DGGML_CCACHE=OFF \
  -DLLAMA_CURL=OFF \
  -DLLAMA_HTTPLIB=OFF \
  -DLLAMA_BUILD_BORINGSSL=OFF \
  -DLLAMA_OPENSSL=OFF \
  -DCMAKE_POSITION_INDEPENDENT_CODE=ON \
  ../llama/llama.cpp

注意：

所有 -D 选项直接对应之前的 --CD... 参数。
我们移除了 --CDCMAKE_CONFIGURATION_TYPES，因为单配置生成器用 CMAKE_BUILD_TYPE 即可。

使用 Conda 的 make 进行编译

(openclaw-build) [root@k8s-master build-linux-x64]# /root/miniconda3/envs/openclaw-build/bin/make -j6

测试 OpenClaw

# 确保 LD_LIBRARY_PATH 仍指向 Conda 的 lib
(openclaw-build) [root@k8s-master build-linux-x64]# export LD_LIBRARY_PATH=/root/miniconda3/envs/openclaw-build/lib:$LD_LIBRARY_PATH
(openclaw-build) [root@k8s-master build-linux-x64]# openclaw --version
OpenClaw 2026.3.8 (3caab92)

运行引导向导

# 完整安装向导 + 安装系统服务
(openclaw-build) [root@k8s-master build-linux-x64]# openclaw onboard --install-daemon
🦞 OpenClaw 2026.3.8 (3caab92) — You had me at 'openclaw gateway start.'

▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
██░▄▄▄░██░▄▄░██░▄▄▄██░▀██░██░▄▄▀██░████░▄▄▀██░███░██
██░███░██░▀▀░██░▄▄▄██░█░█░██░█████░████░▀▀░██░█░█░██
██░▀▀▀░██░█████░▀▀▀██░██▄░██░▀▀▄██░▀▀░█░██░██▄▀▄▀▄██
▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
                  🦞 OPENCLAW 🦞                    
 
┌  OpenClaw onboarding
│
◇  Security ─────────────────────────────────────────────────────────────────────────────────╮
│                                                                                            │
│  Security warning — please read.                                                           │
│                                                                                            │
│  OpenClaw is a hobby project and still in beta. Expect sharp edges.                        │
│  By default, OpenClaw is a personal agent: one trusted operator boundary.                  │
│  This bot can read files and run actions if tools are enabled.                             │
│  A bad prompt can trick it into doing unsafe things.                                       │
│                                                                                            │
│  OpenClaw is not a hostile multi-tenant boundary by default.                               │
│  If multiple users can message one tool-enabled agent, they share that delegated tool      │
│  authority.                                                                                │
│                                                                                            │
│  If you’re not comfortable with security hardening and access control, don’t run           │
│  OpenClaw.                                                                                 │
│  Ask someone experienced to help before enabling tools or exposing it to the internet.     │
│                                                                                            │
│  Recommended baseline:                                                                     │
│  - Pairing/allowlists + mention gating.                                                    │
│  - Multi-user/shared inbox: split trust boundaries (separate gateway/credentials, ideally  │
│    separate OS users/hosts).                                                               │
│  - Sandbox + least-privilege tools.                                                        │
│  - Shared inboxes: isolate DM sessions (`session.dmScope: per-channel-peer`) and keep      │
│    tool access minimal.                                                                    │
│  - Keep secrets out of the agent’s reachable filesystem.                                   │
│  - Use the strongest available model for any bot with tools or untrusted inboxes.          │
│                                                                                            │
│  Run regularly:                                                                            │
│  openclaw security audit --deep                                                            │
│  openclaw security audit --fix                                                             │
│                                                                                            │
│  Must read: https://docs.openclaw.ai/gateway/security                                      │
│                                                                                            │
├────────────────────────────────────────────────────────────────────────────────────────────╯
│
◇  I understand this is personal-by-default and shared/multi-user use requires lock-down. Continue?
│  Yes
│
◇  Onboarding mode
│  QuickStart
│
◇  QuickStart ─────────────────────────╮
│                                      │
│  Gateway port: 18789                 │
│  Gateway bind: Loopback (127.0.0.1)  │
│  Gateway auth: Token (default)       │
│  Tailscale exposure: Off             │
│  Direct to chat channels.            │
│                                      │
├──────────────────────────────────────╯
│
◇  Model/auth provider
│  Qwen
│
◐  Starting Qwen OAuth…...│
◇  Qwen OAuth ─────────────────────────────────────────────────────────────────────────╮
│                                                                                      │
│  Open https://chat.qwen.ai/authorize?user_code=ZxxxW-BX&client=qwen-code to approve  │
│  access.                                                                             │
│  If prompted, enter the code ZGZTW-BX.                                               │
│                                                                                      │
├──────────────────────────────────────────────────────────────────────────────────────╯
◇  Qwen OAuth complete
│
◇  Model configured ─────────────────────────────╮
│                                                │
│  Default model set to qwen-portal/coder-model  │
│                                                │
├────────────────────────────────────────────────╯
│
◇  Provider notes ──────────────────────────────────────────────────────────────────────╮
│                                                                                       │
│  Qwen OAuth tokens auto-refresh. Re-run login if refresh fails or access is revoked.  │
│  Base URL defaults to https://portal.qwen.ai/v1. Override                             │
│  models.providers.qwen-portal.baseUrl if needed.                                      │
│                                                                                       │
├───────────────────────────────────────────────────────────────────────────────────────╯
│
◇  Default model
│  Enter model manually
│##Qwen 为 Qwen Coder 和 Qwen Vision 模型提供免费层 OAuth 流程（每天 2,000 次请求，受 Qwen 速率限制约束）
◇  Default model
│  qwen-portal/coder-model
│
◇  Channel status ────────────────────────────╮
│                                             │
│  Telegram: needs token                      │
│  WhatsApp (default): not linked             │
│  Discord: needs token                       │
│  Slack: needs tokens                        │
│  Signal: needs setup                        │
│  signal-cli: missing (signal-cli)           │
│  iMessage: needs setup                      │
│  imsg: missing (imsg)                       │
│  IRC: not configured                        │
│  Google Chat: not configured                │
│  LINE: not configured                       │
│  Feishu: install plugin to enable           │
│  Google Chat: install plugin to enable      │
│  Nostr: install plugin to enable            │
│  Microsoft Teams: install plugin to enable  │
│  Mattermost: install plugin to enable       │
│  Nextcloud Talk: install plugin to enable   │
│  Matrix: install plugin to enable           │
│  BlueBubbles: install plugin to enable      │
│  LINE: install plugin to enable             │
│  Zalo: install plugin to enable             │
│  Zalo Personal: install plugin to enable    │
│  Synology Chat: install plugin to enable    │
│  Tlon: install plugin to enable             │
│                                             │
├─────────────────────────────────────────────╯
│
◇  How channels work ───────────────────────────────────────────────────────────────────────╮
│                                                                                           │
│  DM security: default is pairing; unknown DMs get a pairing code.                         │
│  Approve with: openclaw pairing approve <channel> <code>                                  │
│  Public DMs require dmPolicy="open" + allowFrom=["*"].                                    │
│  Multi-user DMs: run: openclaw config set session.dmScope "per-channel-peer" (or          │
│  "per-account-channel-peer" for multi-account channels) to isolate sessions.              │
│  Docs: channels/pairing              │
│                                                                                           │
│  Telegram: simplest way to get started — register a bot with @BotFather and get going.    │
│  WhatsApp: works with your own number; recommend a separate phone + eSIM.                 │
│  Discord: very well supported right now.                                                  │
│  IRC: classic IRC networks with DM/channel routing and pairing controls.                  │
│  Google Chat: Google Workspace Chat app with HTTP webhook.                                │
│  Slack: supported (Socket Mode).                                                          │
│  Signal: signal-cli linked device; more setup (David Reagans: "Hop on Discord.").         │
│  iMessage: this is still a work in progress.                                              │
│  LINE: LINE Messaging API webhook bot.                                                    │
│  Feishu: 飞书/Lark enterprise messaging with doc/wiki/drive tools.                        │
│  Nostr: Decentralized protocol; encrypted DMs via NIP-04.                                 │
│  Microsoft Teams: Bot Framework; enterprise support.                                      │
│  Mattermost: self-hosted Slack-style chat; install the plugin to enable.                  │
│  Nextcloud Talk: Self-hosted chat via Nextcloud Talk webhook bots.                        │
│  Matrix: open protocol; install the plugin to enable.                                     │
│  BlueBubbles: iMessage via the BlueBubbles mac app + REST API.                            │
│  Zalo: Vietnam-focused messaging platform with Bot API.                                   │
│  Zalo Personal: Zalo personal account via QR code login.                                  │
│  Synology Chat: Connect your Synology NAS Chat to OpenClaw with full agent capabilities.  │
│  Tlon: decentralized messaging on Urbit; install the plugin to enable.                    │
│                                                                                           │
├───────────────────────────────────────────────────────────────────────────────────────────╯
│
◇  Select channel (QuickStart)
│  Skip for now
Updated ~/.openclaw/openclaw.json
Workspace OK: ~/.openclaw/workspace
Sessions OK: ~/.openclaw/agents/main/sessions
│
◇  Web search ────────────────────────────────────────╮
│                                                     │
│  Web search lets your agent look things up online.  │
│  Choose a provider and paste your API key.          │
│  Docs: https://docs.openclaw.ai/tools/web           │
│                                                     │
├─────────────────────────────────────────────────────╯
│
◇  Search provider
│  Skip for now
│
◇  Skills status ─────────────╮
│                             │
│  Eligible: 3                │
│  Missing requirements: 41   │
│  Unsupported on this OS: 7  │
│  Blocked by allowlist: 0    │
│                             │
├─────────────────────────────╯
│
◇  Configure skills now? (recommended)
│  No
│
◇  Hooks ──────────────────────────────────────────────────────────────────╮
│                                                                          │
│  Hooks let you automate actions when agent commands are issued.          │
│  Example: Save session context to memory when you issue /new or /reset.  │
│                                                                          │
│  Learn more: https://docs.openclaw.ai/automation/hooks                   │
│                                                                          │
├──────────────────────────────────────────────────────────────────────────╯
│
◇  Enable hooks?
│  Skip for now
Config overwrite: /root/.openclaw/openclaw.json (sha256 1f883aa60beabccbfdb30130a587b297df4fb72c26815577529a17cec31c5235 -> c3716feedf08ca3333c31c3c24a00aa88b4dcb1f1477c9deb2240a3a699ccaf5, backup=/root/.openclaw/openclaw.json.bak)
│
◇  Gateway service runtime ────────────────────────────────────────────╮
│                                                                      │
│  QuickStart uses Node for the Gateway service (stable + supported).  │
│                                                                      │
├──────────────────────────────────────────────────────────────────────╯
│
◇  Gateway service already installed
│  Restart
│
◒  Restarting Gateway service…Restarted systemd service: openclaw-gateway.service
◇  Gateway service restarted.
│
◇  
Agents: main (default)
Heartbeat interval: 30m (main)
Session store (main): /root/.openclaw/agents/main/sessions/sessions.json (1 entries)
- agent:main:main (11m ago)
│
◇  Optional apps ────────────────────────╮
│                                        │
│  Add nodes for extra features:         │
│  - macOS app (system + notifications)  │
│  - iOS app (camera/canvas)             │
│  - Android app (camera/canvas)         │
│                                        │
├────────────────────────────────────────╯
│
◇  Control UI ─────────────────────────────────────────────────────────────────────╮
│                                                                                  │
│  Web UI: http://127.0.0.1:18789/                                                 │
│  Web UI (with token):                                                            │
│  http://127.0.0.1:18789/#token=148097c411bd044417ad3a998fdbda9ac99de6b4b5e62d10  │
│  Gateway WS: ws://127.0.0.1:18789                                                │
│  Gateway: reachable                                                              │
│  Docs: https://docs.openclaw.ai/web/control-ui                                   │
│                                                                                  │
├──────────────────────────────────────────────────────────────────────────────────╯
│
◇  Start TUI (best option!) ─────────────────────────────────╮
│                                                            │
│  This is the defining action that makes your agent you.    │
│  Please take your time.                                    │
│  The more you tell it, the better the experience will be.  │
│  We will send: "Wake up, my friend!"                       │
│                                                            │
├────────────────────────────────────────────────────────────╯
│
◇  Token ────────────────────────────────────────────────────────────────────────────────────╮
│                                                                                            │
│  Gateway token: shared auth for the Gateway + Control UI.                                  │
│  Stored in: ~/.openclaw/openclaw.json (gateway.auth.token) or OPENCLAW_GATEWAY_TOKEN.      │
│  View token: openclaw config get gateway.auth.token                                        │
│  Generate token: openclaw doctor --generate-gateway-token                                  │
│  Web UI keeps dashboard URL tokens in memory for the current tab and strips them from the  │
│  URL after load.                                                                           │
│  Open the dashboard anytime: openclaw dashboard --no-open                                  │
│  If prompted: paste the token into Control UI settings (or use the tokenized dashboard     │
│  URL).                                                                                     │
│                                                                                            │
├────────────────────────────────────────────────────────────────────────────────────────────╯
│
◇  How do you want to hatch your bot?
│  Hatch in TUI (recommended)
 openclaw tui - ws://127.0.0.1:18789 - agent main - session main                                       

 session agent:main:main            

# 仅运行配置向导
(openclaw-build) [root@k8s-master build-linux-x64]# openclaw onboard

完成初始配置向导

首次安装后，OpenClaw 会自动启动配置向导，帮助您快速完成初始设置。您也可以手动执行openclaw onboard命令进行配置。

配置项	建议配置
I understand this is powerful and inherently risky. Continue?	选择 Yes
Onboarding mode	选择 QuickStart
Model/auth provider	选择 Skip for now（稍后配置百炼模型）
Filter models by provider	选择 All providers
Default model	选择 Keep current
Select channel (QuickStart)	选择 Skip for now（稍后配置渠道）
Configure skills now? (recommended)	选择 No
Enable hooks?	按空格键选中选项，按回车键进入下一步
How do you want to hatch your bot?	选择 Do this later

检查安装

# 查看 Gateway 状态
(openclaw-build) [root@k8s-master build-linux-x64]# openclaw gateway status
🦞 OpenClaw 2026.3.8 (3caab92) — I've survived more breaking changes than your last three relationships.

│
◇  
Service: systemd (enabled)
File logs: /tmp/openclaw/openclaw-2026-03-10.log
Command: /root/miniconda3/envs/openclaw-build/bin/node /root/miniconda3/envs/openclaw-build/lib/node_modules/openclaw/dist/index.js gateway --port 18789
Service file: ~/.config/systemd/user/openclaw-gateway.service
Service env: OPENCLAW_GATEWAY_PORT=18789

Config (cli): ~/.openclaw/openclaw.json
Config (service): ~/.openclaw/openclaw.json

Gateway: bind=loopback (127.0.0.1), port=18789 (service args)
Probe target: ws://127.0.0.1:18789
Dashboard: http://127.0.0.1:18789/
Probe note: Loopback-only gateway; only local clients can connect.

Runtime: running (pid 71580, state active, sub running, last exit 0, reason 0)
RPC probe: ok

Listening: 127.0.0.1:18789
Troubles: run openclaw status
Troubleshooting: https://docs.openclaw.ai/troubleshooting

# 打开控制面板
(openclaw-build) [root@k8s-master build-linux-x64]# openclaw dashboard

🦞 OpenClaw 2026.3.8 (3caab92) — No $999 stand required.

Dashboard URL: http://127.0.0.1:18789/#token=57fcbb62ae4f354278d50dae53f2a1f9317b86c6a90ba7fa
Copy to clipboard unavailable.
No GUI detected. Open from your computer:
ssh -N -L 18789:127.0.0.1:18789 root@10.10.10.103
Then open:
http://localhost:18789/
http://localhost:18789/#token=57fcbb62ae4f354278d50dae53f2a1f9317b86c6a90ba7fa
Docs:
https://docs.openclaw.ai/gateway/remote
https://docs.openclaw.ai/web/control-ui

⚡快速开始-接入千问百炼模型

1. 启动 Gateway

# 前台运行（适合调试）
(openclaw-build) [root@k8s-master build-linux-x64]# openclaw gateway --port 18789 --verbose

# 使用守护进程（后台运行）
(openclaw-build) [root@k8s-master build-linux-x64]# openclaw gateway start
🦞 OpenClaw 2026.3.8 (3caab92)
   One CLI to rule them all, and one more restart because you changed the port.

Restarted systemd service: openclaw-gateway.service

2. 打开控制面板

(openclaw-build) [root@k8s-master build-linux-x64]# openclaw dashboard
🦞 OpenClaw 2026.3.8 (3caab92) — No $999 stand required.

Dashboard URL: http://127.0.0.1:18789/#token=57fcbb62ae4f354278d50dae53f2a1f9317b86c6a90ba7fa
Copy to clipboard unavailable.
No GUI detected. Open from your computer:
ssh -N -L 18789:127.0.0.1:18789 root@10.10.10.103
Then open:
http://localhost:18789/
http://localhost:18789/#token=57fcbb62ae4f354278d50dae53f2a1f9317b86c6a90ba7fa
Docs:
https://docs.openclaw.ai/gateway/remote
https://docs.openclaw.ai/web/control-ui

在这里插入图片描述

3. 配置阿里百炼

阿里百炼提供100万免费token
地址: https://bailian.console.aliyun.com/cn-beijing/?tab=model#/model-usage/free-quota

在这里插入图片描述

4. 创建API_KEY

点击密钥管理创建

在这里插入图片描述

5. 配置百炼模型

# 打开控制面板
(openclaw-build) [root@k8s-master build-linux-x64]# openclaw dashboard

# windows浏览器访问
win+r打开powershell，输入以下命令和服务器密码后，不要关闭此窗口，通过隧道实现dashboard访问
PS C:\Users\王> ssh -N -L 18789:127.0.0.1:18789 root@10.10.10.103

Authorized users only. All activities may be monitored and reported.
root@10.10.10.103's password:

在这里插入图片描述

6. 点击配置，选择RAW，更新json文件

替换apikey部分，具体的key见第四步。点击save，再点击UPDATE

https://bailian.console.aliyun.com/cn-beijing/?tab=doc#/doc/?type=model&url=3020785

  "models": {
    "mode": "merge",
    "providers": {
      "bailian": {
        "baseUrl": "https://dashscope.aliyuncs.com/compatible-mode/v1",
        "apiKey": "DASHSCOPE_API_KEY",
        "api": "openai-completions",
        "models": [
          {
            "id": "qwen3.5-plus",
            "name": "qwen3.5-plus",
            "reasoning": false,
            "input": ["text", "image"],
            "contextWindow": 1000000,
            "maxTokens": 65536
          },
          {
            "id": "qwen3-coder-next",
            "name": "qwen3-coder-next",
            "reasoning": false,
            "input": ["text"],
            "contextWindow": 262144,
            "maxTokens": 65536
          }
        ]
      }
    }
  },
  "agents": {
    "defaults": {
      "model": {
        "primary": "bailian/qwen3.5-plus"
      },
      "models": {
        "bailian/qwen3.5-plus": {},
        "bailian/qwen3-coder-next": {}
      }
    }
  },

在这里插入图片描述

7. 重启gateway

(openclaw-build) [root@k8s-master build-linux-x64]# openclaw gateway start
🦞 OpenClaw 2026.3.8 (3caab92)
   One CLI to rule them all, and one more restart because you changed the port.

Restarted systemd service: openclaw-gateway.service

8. 对话测试-dashboard控制台

在这里插入图片描述

📁Workspace 工作空间

Workspace 是 OpenClaw 智能体的工作目录，也是它读取上下文、保存记忆、执行工具操作的唯一位置。

Workspace 结构

在这里插入图片描述

核心文件说明

文件	用途	加载时机
`AGENTS.md`	操作指令、连续性规则、重要记忆	每会话首条消息
`SOUL.md`	人格定义、语气、边界	每会话首条消息
`TOOLS.md`	环境特定的工具说明	每会话首条消息
`USER.md`	用户信息、偏好、项目背景	每会话首条消息
`IDENTITY.md`	助手身份（名称、表情、头像）	每会话首条消息
`MEMORY.md`	长期记忆（安全敏感内容）	仅主会话

⚙️配置 openclaw.json

~/.openclaw/openclaw.json 是 OpenClaw 的核心配置文件，使用 JSON5 格式（支持注释和尾随逗号）。

最小配置示例

{
  "agents": {
    "defaults": {
      "workspace": "~/.openclaw/workspace"
    }
  },
  "channels": {
    "telegram": {
      "enabled": true,
      "botToken": "123456:ABC...",
      "dmPolicy": "pairing",
      "allowFrom": ["tg:123456789"]
    }
  }
}

Agents 配置

{
  "agents": {
    "defaults": {
      "workspace": "~/.openclaw/workspace",
      "model": {
        "primary": "anthropic/claude-sonnet-4-5",
        "fallbacks": ["openai/gpt-5.2"]
      },
      "heartbeat": {
        "every": "30m",
        "target": "last"
      }
    }
  }
}

Session 会话管理

{
  "session": {
    "dmScope": "per-channel-peer",
    "reset": {
      "mode": "daily",
      "atHour": 4,
      "idleMinutes": 120
    }
  }
}

💬配置对话终端（Channels）

DM 安全策略

策略	说明	适用场景
`pairing`	未知发送者获得配对码，需主人批准	默认，最安全
`allowlist`	仅允许列表中的发送者	已知联系人
`open`	允许所有入站 DM	公开机器人
`disabled`	忽略所有 DM	仅群组使用

Telegram 配置

{
  "channels": {
    "telegram": {
      "enabled": true,
      "botToken": "your-bot-token",
      "dmPolicy": "pairing",
      "allowFrom": ["tg:123456789"],
      "streaming": "partial"
    }
  }
}

🤖配置多智能体（Multi-Agent）

{
  "agents": {
    "defaults": {
      "workspace": "~/.openclaw/workspace"
    },
    "list": [
      {
        "id": "main",
        "description": "通用助手"
      },
      {
        "id": "coder",
        "workspace": "~/.openclaw/workspace-coder",
        "description": "编程专家"
      }
    ]
  }
}

🧩技能安装

OpenClaw 使用技能系统来扩展功能。技能是可复用的功能模块，可以通过 ClawHub CLI 从 clawhub.com 安装。

💡

技能让你可以一键添加新能力，如图像生成、天气查询、GitHub 操作、视频处理等，无需手动配置每个工具。

安装 ClawHub CLI

首先确保已安装 ClawHub 命令行工具：

npm install -g clawhub

搜索技能

在 ClawHub 上搜索你需要的技能：

# 搜索关键词
clawhub search "postgres backups"
clawhub search "image generation"
clawhub search "weather"

安装技能

找到想要的技能后，使用以下命令安装：

# 安装最新版本
clawhub install baoyu-image-gen

# 安装指定版本
clawhub install baoyu-image-gen --version 1.2.3

管理技能

命令	说明	示例
`clawhub list`	列出已安装的技能	查看当前工作空间的所有技能
`clawhub update <skill>`	更新指定技能到最新版本	`clawhub update baoyu-image-gen`
`clawhub update --all`	更新所有技能	批量更新所有已安装技能
`clawhub update --force`	强制更新（忽略版本检查）	解决版本冲突时使用

常用推荐技能

技能名称	功能	安装命令
`baoyu-image-gen`	AI 图像生成（OpenAI、Google、Replicate 等）	`clawhub install baoyu-image-gen`
`weather`	天气查询和预报	`clawhub install weather`
`github`	GitHub 操作（Issues、PR、代码审查）	`clawhub install github`
`video-frames`	视频帧提取和剪辑	`clawhub install video-frames`
`xurl`	X (Twitter) API 操作	`clawhub install xurl`
`find-skills`	帮助发现和安装技能	`clawhub install find-skills`

技能工作原理

技能是包含以下内容的文件夹：

my-skill/├── SKILL.md # 技能定义和使用说明└── 其他文件 # 脚本、配置等

✅

安装后，OpenClaw 会自动识别技能能力，并在相关任务触发时自动调用。无需额外配置。

发布自己的技能

如果你有自己开发的技能，可以发布到 ClawHub 分享：

# 登录 ClawHub
clawhub login

# 发布技能
clawhub publish ./my-skill \
  --slug my-skill \
  --name "My Skill" \
  --version 1.0.0 \
  --changelog "Initial release"

⚠️

**注意：**技能默认注册表为 https://clawhub.com，可通过 CLAWHUB_REGISTRY 环境变量或 --registry 参数覆盖。

⌨️常用 CLI 命令

Gateway 管理

# 查看状态
openclaw gateway status

# 启动/停止/重启
openclaw gateway start
openclaw gateway stop
openclaw gateway restart

配置管理

# 运行配置向导
openclaw onboard

# 获取/设置配置值
openclaw config get agents.defaults.workspace
openclaw config set agents.defaults.model.primary "openai/gpt-5.2"

诊断工具

# 健康检查
openclaw doctor

# 自动修复
openclaw doctor --fix

# 查看日志
openclaw logs --follow

🔒安全最佳实践

⚠️

永远不要在未经保护的情况下公开 DM：
使用 dmPolicy: "pairing" 或 allowlist 而非 open

多用户环境

{
  "session": {
    "dmScope": "per-channel-peer"
  }
}

沙盒执行

{
  "agents": {
    "defaults": {
      "sandbox": {
        "mode": "non-main",
        "scope": "agent"
      }
    }
  }
}

🔧故障排除

错误	原因	解决
`Config validation failed`	配置格式错误	运行 `openclaw doctor`
`Unauthorized`	API Key 无效	检查 `auth` 配置
`Session not found`	会话已过期	发送 `/new` 重置

🔗资源链接

官网: openclaw.ai
文档: docs.openclaw.ai
GitHub: github.com/openclaw/openclaw
Discord: discord.gg/clawd
技能市场: clawhub.com

📝总结

OpenClaw 是一个功能强大、灵活的个人 AI 助手平台。关键要点：

Gateway 是核心 - 所有功能都围绕 Gateway 展开
Workspace 是工作空间 - 保持文件整洁，控制 token 使用
配置即代码 - openclaw.json 定义一切行为
安全第一 - 默认使用配对模式，谨慎开放 DM

🔧故障排除

错误	原因	解决
`Config validation failed`	配置格式错误	运行 `openclaw doctor`
`Unauthorized`	API Key 无效	检查 `auth` 配置
`Session not found`	会话已过期	发送 `/new` 重置