Openclaw 实践笔记
·
目录
说明
- 在我的香橙派上安装小龙虾，并在飞书上喊它为我做事情
- 模型切换
- 全流程测试
1 搭建Openclaw
1.1 预备工作
- 供应商模型及API Keys：
| 厂商/条项 | 百炼千问 | 智谱 | Kimi | 超算互联网 |
|---|---|---|---|---|
| 开放平台 | https://bailian.console.aliyun.com | https://open.bigmodel.cn | https://platform.moonshot.cn | https://www.scnet.cn/ui/mall/ |
| 模型 | qwen-turbo, qwen3.5-plus .... | glm-5 | kimi-k2.5 | MiniMax-M2.5 |
| Baseurl | https://dashscope.aliyuncs.com/compatible-mode/v1 | https://open.bigmodel.cn/api/paas/v4 | https://api.moonshot.cn/v1 | https://api.scnet.cn/api/llm/v1 |
| ApiKeys | -- | -- | -- | -- |
- Channel供应商
| 厂商/条项 | 飞书 | |
|---|---|---|
| 开放平台 | https://open.feishu.cn | |
| app id | 略，（一般为 cli_xxx, 平台里面建项目应用后获取） | |
| app secret | 略 |
- 系统配置
香橙派5B（8核8G+128TF卡），Ubuntu22.04
1.2 安装
注：用root登录，而不是其他用户登陆后用su，不然会安装服务失败，只能手动安装。
- 采用npm安装openclaw方式
注：安装channels时先略过不配置（目前安装有点小bug，会触发重复配置的错误），等装完系统后再通过chat交互界面来自动设置。
(https://docs.openclaw.ai/install#npm-pnpm)
1. Remove any existing Node.js installations (optional but recommended):
sudo apt purge nodejs npm -y
sudo apt autoremove -y
2. Install NVM (Node Version Manager):
sudo apt update
sudo apt install curl -y
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.39.7/install.sh | bash
3. Activate NVM in your current session:
export NVM_DIR="$HOME/.nvm"
[ -s "$NVM_DIR/nvm.sh" ] && \. "$NVM_DIR/nvm.sh" # This loads nvm
[ -s "$NVM_DIR/bash_completion" ] && \. "$NVM_DIR/bash_completion" # This loads nvm bash_completion
4. Install Node.js v22 (or the latest version):
nvm install node
# Or to install a specific version, e.g., 22:
nvm install 22
npm install -g openclaw@latest
openclaw onboard --install-daemon
# 安装服务
systemctl start user@$(id -u)
systemctl --user daemon-reload
systemctl --user enable openclaw-gateway.service
systemctl --user start openclaw-gateway.service
- 局域网访问以及允许跨域
"bind": "lan",
"controlUi": {
"allowedOrigins": [
"http://172.16.1.20:18789",
"http://localhost:18789",
"http://127.0.0.1:18789"
],
"allowInsecureAuth": true,
"dangerouslyDisableDeviceAuth": true安装完成后，管理后台可通过http://{ip}:18789/#token=xxx访问。
- 连接飞书
可在管理后台的聊天界面通过对话方式让小龙虾与你的飞书群连接，它会自动安装飞书扩展并让你提供对应的app id等信息帮助设置。
步骤：
- 先在飞书建立应用openclaw-chat，然后添加机器人，添加相关权限
- 事件订阅方式采用长连接，添加相关事件
- 在飞书应用建群，在群配置-添加群机器人openclaw-chat
- 首次与飞书连接时，需要在飞书对话里获取验证码，然后在终端输入配对命令"openclaw pairing approve feishu {code}"审批通过，如图：
- 效果
2 模型配置与切换
- 执行命令
openclaw models set <模型 ID>
注：有时修改后，有些会话所用的model并不能及时切换（即使restart gateway），这时候需要重启机器再观察。
- 配置主文件位于 ~/.openclaw/openclaw.json, 有两个需要定义的地方, 一个是models/provider节, 一个是agents/models节, 如:
# 配置模型供应详细
"models": {
"providers": {
"scnet": {
"baseUrl": "https://api.scnet.cn/api/llm/v1",
"apiKey": "...",
"api": "openai-completions",
"models": [
{
"id": "MiniMax-M2.5",
"name": "MiniMax-M2.5",
"api": "openai-completions",
"reasoning": false,
"input": [
"text",
"image"
],
"cost": {
"input": 0,
"output": 0,
"cacheRead": 0,
"cacheWrite": 0
},
"contextWindow": 1000000,
"maxTokens": 65536
},
...
# 定义需要使用的models
"agents": {
"defaults": {
"model": {
"primary": "bailian/qwen-turbo"
},
"models": {
"scnet/MiniMax-M2.5": {},
"scnet/DeepSeek-R1-0528": {},
"qwen-portal/coder-model": {
"alias": "qwen"
},
- 完整参考：
{
meta: {
lastTouchedVersion: '2026.3.28',
lastTouchedAt: '2026-04-01T03:36:31.935Z',
},
wizard: {
lastRunAt: '2026-03-31T03:40:30.850Z',
lastRunVersion: '2026.3.28',
lastRunCommand: 'onboard',
lastRunMode: 'local',
},
auth: {
profiles: {
'moonshot:default': {
provider: 'moonshot',
mode: 'api_key',
},
},
},
models: {
mode: 'merge',
providers: {
moonshot: {
baseUrl: 'https://api.moonshot.cn/v1',
api: 'openai-completions',
models: [
{
id: 'kimi-k2.5',
name: 'Kimi K2.5',
reasoning: false,
input: [
'text',
'image',
],
cost: {
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
contextWindow: 262144,
maxTokens: 262144,
},
],
},
bailian: {
baseUrl: 'https://dashscope.aliyuncs.com/compatible-mode/v1',
apiKey: '__OPENCLAW_REDACTED__',
api: 'openai-completions',
models: [
{
id: 'qwen-turbo',
name: 'qwen-turbo',
api: 'openai-completions',
reasoning: false,
input: [
'text',
'image',
],
cost: {
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
contextWindow: 1000000,
maxTokens: 65536,
},
{
id: 'qwen3.5-plus',
name: 'qwen3.5-plus',
api: 'openai-completions',
reasoning: false,
input: [
'text',
'image',
],
cost: {
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
contextWindow: 1000000,
maxTokens: 65536,
},
{
id: 'qwen3-coder-next',
name: 'qwen3-coder-next',
api: 'openai-completions',
reasoning: false,
input: [
'text',
],
cost: {
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
contextWindow: 262144,
maxTokens: 65536,
},
],
},
zhipu: {
baseUrl: 'https://open.bigmodel.cn/api/paas/v4',
apiKey: '__OPENCLAW_REDACTED__',
api: 'openai-completions',
models: [
{
id: 'glm-5',
name: 'GLM 5.0',
api: 'openai-completions',
reasoning: true,
input: [
'text',
],
cost: {
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
contextWindow: 128000,
maxTokens: 8192,
},
{
id: 'glm-5-flash',
name: 'GLM 5.0 Flash',
api: 'openai-completions',
reasoning: false,
input: [
'text',
],
cost: {
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
contextWindow: 128000,
maxTokens: 8192,
},
],
},
scnet: {
baseUrl: 'https://api.scnet.cn/api/llm/v1',
apiKey: '__OPENCLAW_REDACTED__',
api: 'openai-completions',
models: [
{
id: 'MiniMax-M2.5',
name: 'MiniMax-M2.5',
api: 'openai-completions',
reasoning: false,
input: [
'text',
'image',
],
cost: {
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
contextWindow: 128000,
maxTokens: 65536,
},
{
id: 'DeepSeek-R1-0528',
name: 'DeepSeek-R1-0528',
api: 'openai-completions',
reasoning: false,
input: [
'text',
'image',
],
cost: {
input: 0,
output: 0,
cacheRead: 0,
cacheWrite: 0,
},
contextWindow: 128000,
maxTokens: 65536,
},
],
},
},
},
agents: {
defaults: {
model: {
primary: 'moonshot/kimi-k2.5',
},
models: {
'moonshot/kimi-k2.5': {
alias: 'Kimi',
},
'bailian/qwen3.5-plus': {},
'bailian/qwen-turbo': {},
'bailian/qwen3-coder-next': {},
'zhipu/glm-5': {},
'zhipu/glm-5-flash': {},
'scnet/MiniMax-M2.5': {
alias: 'MiniMax',
},
'scnet/DeepSeek-R1-0528': {
alias: 'DeepSeek-R1',
},
},
workspace: '/root/.openclaw/workspace',
},
},
tools: {
profile: 'coding',
web: {
search: {
provider: 'kimi',
},
},
},
commands: {
native: 'auto',
nativeSkills: 'auto',
restart: true,
ownerDisplay: 'raw',
},
session: {
dmScope: 'per-channel-peer',
},
hooks: {
internal: {
enabled: true,
entries: {
'command-logger': {
enabled: true,
},
'boot-md': {
enabled: true,
},
},
},
},
channels: {
feishu: {
enabled: true,
appId: 'cli_a93bf33765f81bc0',
appSecret: '__OPENCLAW_REDACTED__',
connectionMode: 'websocket',
domain: 'feishu',
groupPolicy: 'open',
webhookPath: '/feishu/events',
dmPolicy: 'pairing',
reactionNotifications: 'own',
typingIndicator: true,
resolveSenderNames: true,
},
},
gateway: {
port: 18789,
mode: 'local',
bind: 'lan',
controlUi: {
allowedOrigins: [
'http://172.16.1.20:18789',
'http://localhost:18789',
'http://127.0.0.1:18789',
],
allowInsecureAuth: true,
dangerouslyDisableDeviceAuth: true,
},
auth: {
mode: 'token',
token: '__OPENCLAW_REDACTED__',
},
tailscale: {
mode: 'off',
resetOnExit: false,
},
nodes: {
denyCommands: [
'camera.snap',
'camera.clip',
'screen.record',
'contacts.add',
'calendar.add',
'reminders.add',
'sms.send',
],
},
},
skills: {
install: {
nodeManager: 'pnpm',
},
entries: {
'coding-agent': {
enabled: true,
},
discord: {
enabled: false,
},
},
},
plugins: {
entries: {
feishu: {
enabled: true,
config: {},
},
},
},
}3 Openclaw常用命令
# openclaw 医生
openclaw doctor
# 配置入口
openclaw configure --section gateway
# 重启入口网关
openclaw gateway restart
# 配置LLM访问验证
openclaw models auth login --provider qwen-portal
# 更改模型
openclaw models set <模型 ID>
# 查会话列表
openclaw sessions
# 查模型
openclaw models
openclaw models list
# channel或飞书相关
openclaw channels logs # 实时查看日志，确认是否出现 "not in groupAllowFrom" 错误
openclaw config get channels.feishu.appId # 获取飞书应用id
openclaw config get channels.feishu.appSecret
openclaw config get channels.feishu.groupPolicy
openclaw config get channels.feishu.groupAllowFrom4 错误记录
- API rate limit reached. Please try again later.
A: 厂商模型API调用配额满
- "Gateway install failed: Error: systemctl enable failed: Failed to enable unit: Unit file openclaw-gateway.service does not exist."
A:
# 建立服务文件
cat > ~/.config/systemd/user/openclaw-gateway.service <<'EOF'
[Unit]
Description=OpenClaw Gateway
After=network-online.target
Wants=network-online.target
[Service]
Type=simple
ExecStart=/root/.nvm/versions/node/v25.8.2/bin/openclaw gateway --port 18789
Restart=always
RestartSec=5
Environment=HOME=/home/$USER
Environment=PATH=/home/$USER/.npm-global/bin:/usr/local/bin:/usr/bin:/bin
[Install]
WantedBy=default.target
EOF
# 安装服务
systemctl start user@$(id -u)
systemctl --user daemon-reload
systemctl --user enable openclaw-gateway.service
systemctl --user start openclaw-gateway.service5 参考
- 官网 openclaw.ai
- 部署百炼千问 https://blog.csdn.net/qq_37212162/article/details/158890174
AtomGit 是由开放原子开源基金会联合 CSDN 等生态伙伴共同推出的新一代开源与人工智能协作平台。平台坚持“开放、中立、公益”的理念，把代码托管、模型共享、数据集托管、智能体开发体验和算力服务整合在一起，为开发者提供从开发、训练到部署的一站式体验。
更多推荐
4
0- 0
已为社区贡献1条内容
所有评论(0)