Token导航 LogoToken导航

Agent Skill · 研究检索

openclaw-cli(OpenClaw CLI 命令行)

openclaw-cli 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 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。

openclaw-cli 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为操作浏览器;许可证信息还不明确,第三方扫描有未通过项,按安装量在研究检索类的 73,675 个 Skill 里排第 17,636。

总安装 6,492周安装 276GitHub Stars 公开资料未说明下载 2,274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
unknown

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:openclaw-cli(OpenClaw CLI 命令行)
来源仓库:https://github.com/irangareddy/openclaw-essentials
仓库路径:skills/openclaw-cli
安装命令:
npx skills add https://github.com/irangareddy/openclaw-essentials --skill openclaw-cli
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 npx skills 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。

skills.shnpx skills
npx skills add https://github.com/irangareddy/openclaw-essentials --skill openclaw-cli

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。

openclaw-cli 用于查找、检索和筛选相关信息,适合快速定位候选结果。适用于需要根据关键词或任务场景从来源线索中筛选信息的场景。通过 npx skills add 命令安装,需结合原始 README 核验具体用法。安装前建议确认权限范围、维护状态及是否会触发联网或文件读写操作。

适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/openclaw-cli

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 Codex、Claude、Cursor、Gemini CLI;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 npx skills add https://github.com/irangareddy/openclaw-essentials --skill opencl…;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

    它需要操作浏览器,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

openclaw-cli 怎么安装?

在终端执行 npx skills add https://github.com/irangareddy/openclaw-essentials --skill openclaw-cli。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

openclaw-cli 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。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。

openclaw-cli 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

openclaw-cli 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

openclaw-cli 用的人多吗?

公开数据是累计安装 6,492 次,近一周安装 276 次,下载 2,274 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。

openclaw-cli 安全吗?

第三方扫描有未通过项,风险较高。除非你确认过源码,否则不建议直接安装到常用环境。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 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 Codex,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。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 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
openclaw-cli

OpenClaw CLI 命令行 · 本页

6,492—操作浏览器未明确未通过

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

OpenClaw CLI

Complete reference for openclaw command-line interface operations.

When to Use

Managing OpenClaw gateway, agents, channels, skills, hooks, and automation.

Core Commands

Setup & Onboarding

Initial setup:

# Quick onboarding with daemon install
openclaw onboard --install-daemon

# Setup workspace and config
openclaw setup --workspace ~/.openclaw/workspace

# Interactive configuration
openclaw configure

Health check:

openclaw doctor

Gateway Management

Run gateway:

# Interactive mode
openclaw gateway

# With specific port
openclaw gateway --port 18789

# With tailscale
openclaw gateway --tailscale serve

Gateway service:

# Install as system service
openclaw gateway install

# Control service
openclaw gateway start
openclaw gateway stop
openclaw gateway restart
openclaw gateway status

Gateway health:

openclaw status
openclaw status --deep     # Probe channels
openclaw health

Agent Management

List agents:

openclaw agents list
openclaw agents list --bindings    # Show routing

Add agent:

# Interactive wizard
openclaw agents add <name>

# Non-interactive
openclaw agents add dev \
  --workspace ~/.openclaw/workspace-dev \
  --model claude-sonnet-4.5 \
  --non-interactive

Delete agent:

openclaw agents delete <id>

Set identity:

# From IDENTITY.md
openclaw agents set-identity --agent main --from-identity

# Explicit values
openclaw agents set-identity --agent main \
  --name "MyAgent" \
  --emoji "🤖" \
  --avatar avatars/bot.png

Skills Management

List skills:

openclaw skills list
openclaw skills list --eligible    # Only ready skills
openclaw skills list --json        # JSON output

Skill info:

openclaw skills info <skill-name>

Check eligibility:

openclaw skills check

Hooks Management

List hooks:

openclaw hooks list
openclaw hooks list --eligible
openclaw hooks list --verbose      # Show missing requirements

Hook info:

openclaw hooks info <hook-name>

Enable/disable:

openclaw hooks enable session-memory
openclaw hooks disable command-logger

Install hooks:

# From npm
openclaw hooks install @openclaw/my-hooks

# Local directory
openclaw hooks install ./my-hooks

# Link (development)
openclaw hooks install -l ./my-hooks

Update hooks:

openclaw hooks update <id>
openclaw hooks update --all

Channel Management

List channels:

openclaw channels list

Channel status:

openclaw channels status
openclaw channels status --probe

Add channel:

# Interactive
openclaw channels add

# Telegram bot
openclaw channels add --channel telegram \
  --account alerts \
  --name "Alerts Bot" \
  --token $TELEGRAM_BOT_TOKEN

# Discord
openclaw channels add --channel discord \
  --account work \
  --token $DISCORD_BOT_TOKEN

Remove channel:

openclaw channels remove --channel telegram --account alerts
openclaw channels remove --channel discord --account work --delete

WhatsApp login:

openclaw channels login --channel whatsapp

Channel logs:

openclaw channels logs
openclaw channels logs --channel whatsapp --lines 100

Models & Authentication

Status:

openclaw models status
openclaw models status --probe               # Live check
openclaw models status --probe-provider anthropic

List models:

openclaw models list
openclaw models list --all
openclaw models list --provider anthropic

Set default:

openclaw models set claude-sonnet-4.5
openclaw models set-image claude-sonnet-4.5

Auth setup:

# Anthropic (recommended)
claude setup-token
openclaw models auth setup-token --provider anthropic

# Or paste token
openclaw models auth paste-token --provider anthropic

Fallbacks:

openclaw models fallbacks list
openclaw models fallbacks add claude-opus-4.6
openclaw models fallbacks remove claude-haiku-4.5
openclaw models fallbacks clear

Scan for models:

openclaw models scan
openclaw models scan --set-default

Messaging

Send message:

openclaw message send \
  --target +15555550123 \
  --message "Hello from OpenClaw"

# Discord channel
openclaw message send \
  --channel discord \
  --target channel:123456 \
  --message "Deployment complete"

Send poll:

openclaw message poll \
  --channel discord \
  --target channel:123 \
  --poll-question "Lunch?" \
  --poll-option "Pizza" \
  --poll-option "Sushi"

Other message operations:

openclaw message read --target +15555550123
openclaw message react --target <id> --emoji "👍"
openclaw message edit --target <id> --message "Updated"
openclaw message delete --target <id>

Browser Control

Status & control:

openclaw browser status
openclaw browser start
openclaw browser stop
openclaw browser tabs

Navigate:

openclaw browser open https://example.com
openclaw browser navigate https://example.com --target-id <id>

Interact:

openclaw browser click "#submit-button"
openclaw browser type "#email" "user@example.com"
openclaw browser press Enter

Capture:

openclaw browser screenshot
openclaw browser screenshot --full-page
openclaw browser snapshot --format ai

Profiles:

openclaw browser profiles
openclaw browser create-profile --name dev
openclaw browser delete-profile --name old

Nodes & Devices

List nodes:

openclaw nodes list
openclaw nodes status --connected

Node operations:

# Describe node
openclaw nodes describe --node <id>

# Run command on node
openclaw nodes run --node <id> --cwd /path -- ls -la

# Notify (macOS)
openclaw nodes notify --node <id> \
  --title "Build Complete" \
  --body "Success" \
  --sound default

Camera:

openclaw nodes camera list --node <id>
openclaw nodes camera snap --node <id> --facing front
openclaw nodes camera clip --node <id> --duration 10s

Canvas:

openclaw nodes canvas snapshot --node <id>
openclaw nodes canvas present --node <id> --target index.html
openclaw nodes canvas hide --node <id>

Screen recording:

openclaw nodes screen record --node <id> --duration 30s

System Commands

System event:

openclaw system event --text "Deployment complete" --mode now

Heartbeat:

openclaw system heartbeat last
openclaw system heartbeat enable
openclaw system heartbeat disable

Presence:

openclaw system presence

Cron Jobs

List jobs:

openclaw cron list
openclaw cron list --all
openclaw cron status

Add job:

# System event every hour
openclaw cron add \
  --name "hourly-check" \
  --every "1h" \
  --system-event "Hourly check"

# Message at specific time
openclaw cron add \
  --name "morning-reminder" \
  --at "09:00" \
  --message "Good morning!"

Manage jobs:

openclaw cron enable <id>
openclaw cron disable <id>
openclaw cron rm <id>
openclaw cron run <id>

Job runs:

openclaw cron runs --id <id> --limit 10

Configuration

Get/set config:

# Get value
openclaw config get agents.defaults.model.primary

# Set value
openclaw config set agents.defaults.model.primary "claude-sonnet-4.5"

# Unset value
openclaw config unset some.config.path

Memory Operations

Memory status:

openclaw memory status

Index memory:

openclaw memory index

Search memory:

openclaw memory search "GraphQL implementation patterns"

Logs

Tail logs:

openclaw logs
openclaw logs --follow
openclaw logs --limit 200
openclaw logs --json

Sandbox

List sandboxes:

openclaw sandbox list

Recreate sandbox:

openclaw sandbox recreate

Security

Security audit:

openclaw security audit
openclaw security audit --deep
openclaw security audit --fix

Plugins

List plugins:

openclaw plugins list
openclaw plugins list --json

Plugin info:

openclaw plugins info <plugin-id>

Install plugin:

openclaw plugins install <path-or-spec>

Enable/disable:

openclaw plugins enable <id>
openclaw plugins disable <id>

Plugin health:

openclaw plugins doctor

Update & Maintenance

Update OpenClaw:

openclaw update

Reset config:

openclaw reset --scope config
openclaw reset --scope config+creds+sessions
openclaw reset --scope full

Uninstall:

openclaw uninstall --service
openclaw uninstall --state
openclaw uninstall --workspace
openclaw uninstall --all

Global Flags

Available everywhere:

--dev               # Use ~/.openclaw-dev for isolation
--profile <name>    # Use ~/.openclaw-<name>
--no-color         # Disable ANSI colors
--json             # Machine-readable output
-V, --version      # Show version

Common Workflows

First-Time Setup

# 1. Onboard with daemon
openclaw onboard --install-daemon

# 2. Pair WhatsApp (or other channel)
openclaw channels login

# 3. Start gateway
openclaw gateway

# 4. Test with message
openclaw message send --target +1234567890 --message "Test"

Multi-Agent Setup

# 1. Add agent
openclaw agents add work --workspace ~/.openclaw/workspace-work

# 2. Set identity
openclaw agents set-identity --agent work --from-identity

# 3. Add binding (in openclaw.json)
# bindings: [{ agentId: "work", match: { channel: "discord" } }]

# 4. List to verify
openclaw agents list --bindings

Hook Automation

# 1. Enable session memory hook
openclaw hooks enable session-memory

# 2. Enable command logger
openclaw hooks enable command-logger

# 3. Verify
openclaw hooks check

# 4. Restart gateway
openclaw gateway restart

Channel Setup

# 1. Add Telegram bot
openclaw channels add --channel telegram \
  --account alerts \
  --token $TELEGRAM_BOT_TOKEN

# 2. Verify
openclaw channels status

# 3. Send test message
openclaw message send --channel telegram \
  --target <chat-id> \
  --message "Bot online"

Model Configuration

# 1. Setup auth
claude setup-token

# 2. Set default model
openclaw models set claude-sonnet-4.5

# 3. Add fallbacks
openclaw models fallbacks add claude-opus-4.6
openclaw models fallbacks add claude-haiku-4.5

# 4. Verify
openclaw models status

Debugging

Check gateway status:

openclaw status --deep
openclaw doctor
openclaw health

View logs:

openclaw logs --follow
openclaw channels logs --lines 200

Test channel:

openclaw channels status --probe

Check skills/hooks:

openclaw skills check
openclaw hooks check
openclaw plugins doctor

Tips

  1. Use --json for scripting - All commands support JSON output
  2. Profile isolation - Use --profile for testing without affecting main config
  3. Doctor fixes - Run openclaw doctor regularly to catch issues
  4. Logs location - ~/.openclaw/logs/ for file logs
  5. Config location - ~/.openclaw/openclaw.json
  6. Workspace - ~/.openclaw/workspace (or custom path)

Resources

平台分布

Codex

35.76%
按下载量换算813

Claude

28.2%
按下载量换算641

Cursor

18.2%
按下载量换算414

Gemini CLI

8.83%
按下载量换算201

安全审计

Gen Agent Trust Hub

可疑

Socket

通过

Snyk

未通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。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