Token导航 LogoToken导航

Agent Skill · 研究检索

operator(operator 搜索)

operator 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。

operator 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 14,196。

总安装 8,753周安装 376GitHub Stars 1下载 3,068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
1.0.0

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:operator(operator 搜索)
来源仓库:https://github.com/promptrotator/operator
安装命令:
openclaw skills install operator
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 operator

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。

operator 用于查找、检索和筛选相关信息,适合在 OpenClaw 中根据关键词快速定位候选结果时使用。适用于管理 AI 代理实例 Operator 队列,包括创建、监控和发送消息。处理身份验证和实例生命周期,支持跨环境运营管理。

安装命令为 openclaw skills install operator,建议确认权限范围和维护状态。需注意是否会触发联网或文件读写,结合来源仓库进一步核验使用细节。

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 operator;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

operator 怎么安装?

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

operator 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。

operator 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。

operator 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。

operator 用的人多吗?

公开数据是累计安装 8,753 次,近一周安装 376 次,下载 3,068 次,GitHub 1 星。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。

operator 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
operator

operator 搜索 · 本页

8,7531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
operator
description
Manage your Operator fleet of AI agent instances. Create, configure, monitor, message, and manage OpenClaw agents. Handles authentication, instance lifecycle, secrets, automations, and webhooks.

Operator Fleet Manager

This skill lets you manage your Operator fleet of OpenClaw agent instances by sending requests to the Operator chat API.

When to Use This Skill

Use this skill when the user:

  • Asks about their Operator instances or agents
  • Wants to create, delete, restart, or configure instances
  • Needs to check instance logs or status
  • Wants to message a running agent
  • Asks about secrets, automations, or webhooks
  • Mentions "Operator", "fleet", or "OpenClaw"

Authentication

Step 1: Check for existing API key

Read the config file to check if the user is already logged in:

cat ~/.operator/config.json 2>/dev/null

Look for operatorApiKey (starts with ck_) and operatorAppUrl in the JSON. If the file exists and has a key, skip to "Using the API" below.

Step 2: Login (if no key found)

If no API key is found, guide the user through browser-based login:

  1. Generate a session ID:
python3 -c "import uuid; print(uuid.uuid4())"
  1. Tell the user to open this URL in their browser (replace SESSION_ID with the generated UUID):
https://www.operator.io/auth/cli?session=SESSION_ID
  1. After the user confirms they've logged in, poll for the API key:
curl -s "https://www.operator.io/api/cli/poll?session=SESSION_ID"

The response will contain operatorApiKey when auth is complete.

  1. Save the credentials:
mkdir -p ~/.operator
python3 -c "
import json
config = {'operatorApiKey': 'THE_KEY', 'operatorAppUrl': 'https://www.operator.io'}
with open('$HOME/.operator/config.json', 'w') as f:
    json.dump(config, f, indent=2)
print('Saved credentials to ~/.operator/config.json')
"

Using the API

Read credentials

OPERATOR_KEY=$(python3 -c "import json; c=json.load(open('$HOME/.operator/config.json')); print(c.get('operatorApiKey',''))")
OPERATOR_URL=$(python3 -c "import json; c=json.load(open('$HOME/.operator/config.json')); print(c.get('operatorAppUrl','https://www.operator.io'))")

Health check

Verify auth is working:

curl -s "$OPERATOR_URL/api/cli/health" \
  -H "Authorization: Bearer $OPERATOR_KEY"

Returns authenticated, email, and planName.

Send a request to the Operator manager

The Operator manager is an AI that has tools for managing your entire fleet. Send it natural language requests and parse the SSE response to extract text and tool results:

curl -sN "$OPERATOR_URL/api/chat" \
  -H "Authorization: Bearer $OPERATOR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"messages":[{"role":"user","parts":[{"type":"text","text":"YOUR_MESSAGE_HERE"}]}]}' \
  | python3 -c "
import sys, json
for line in sys.stdin:
    line = line.strip()
    if not line.startswith('data: '):
        continue
    try:
        event = json.loads(line[6:])
        t = event.get('type','')
        if t == 'text':
            print(event.get('value',''), end='')
        elif t == 'tool-result':
            result = event.get('result')
            if result is not None:
                print(json.dumps(result, indent=2))
        elif t == 'error':
            print('ERROR:', event.get('value',''))
    except:
        pass
print()
"

To continue a conversation (for follow-up requests), include the id field. The chat ID is returned in the stream's start event:

curl -sN "$OPERATOR_URL/api/chat" \
  -H "Authorization: Bearer $OPERATOR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"id":"CHAT_ID","messages":[{"role":"user","parts":[{"type":"text","text":"YOUR_FOLLOWUP"}]}]}' \
  | python3 -c "
import sys, json
for line in sys.stdin:
    line = line.strip()
    if not line.startswith('data: '):
        continue
    try:
        event = json.loads(line[6:])
        t = event.get('type','')
        if t == 'text':
            print(event.get('value',''), end='')
        elif t == 'tool-result':
            result = event.get('result')
            if result is not None:
                print(json.dumps(result, indent=2))
        elif t == 'error':
            print('ERROR:', event.get('value',''))
    except:
        pass
print()
"

Understanding the SSE response

The API returns a Server-Sent Events stream. Each line is data: {JSON}. The important event types are:

  • {"type":"start"} — stream started
  • {"type":"start-step"} — the manager is beginning a step (may use tools)
  • {"type":"text","value":"..."} — text response chunks (concatenate for full answer)
  • {"type":"tool-call","toolCallId":"...","toolName":"...","args":{}} — the manager is calling a tool
  • {"type":"tool-result","toolCallId":"...","result":{}} — tool result data (instance lists, config, logs, etc.)
  • {"type":"finish-step"} — step complete
  • {"type":"finish","finishReason":"stop"} — response complete

The python pipe above extracts just the text and tool results, giving you clean readable output. Always use this pipe when calling the API.

What the Operator Manager Can Do

The manager has tools for:

  • Instances: list, create, delete, restart, clone, get details, check capacity, get logs
  • Config: update instance configuration (JSON patch), deploy skills, list/read/write workspace files
  • Agents: message a running agent, check latest session activity
  • Secrets: list user secrets, grant/revoke instance access
  • Automations: list, create, update, delete scheduled cron automations
  • Webhooks: list, create, update, delete webhook triggers
  • Checkpoints: search and install agent checkpoint repos

Important Notes

  • Always read credentials from ~/.operator/config.json before every API call.
  • If a 401 error is returned, the API key may be expired. Run the login flow again.
  • If a 429 error is returned, the user has hit their rate limit. Wait before retrying.
  • The manager handles multi-step operations internally. Send one natural language request and it will use its tools as needed.
  • Never expose the raw API key value to the user. Show only a masked version like ck_...xxxx.

平台分布

OpenClaw

72.64%
按下载量换算2,229

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