← 返回 Skills 市场

Book Window Cleaning

Name: Book Window Cleaning
Author: edwardrodriguez703-design

作者 Lokuli · GitHub ↗ · v1.0.1

cross-platform ✓ 安全检测通过

1391

总下载

当前安装

版本数

在 OpenClaw 中安装

/install book-window-cleaning

功能描述

使用说明 (SKILL.md)

uook window cleaning

Book window-cleaning services through Lokuli's MCP server.

MCP Endpoint

https://lokuli.com/mcp/sse

Transport: SSE | JSON-RPC 2.0 | POST requests

Tools

search

{
  "method": "tools/call",
  "params": {
    "name": "search",
    "arguments": {
      "query": "window-cleaning",
      "zipCode": "90640",
      "maxResults": 20
    }
  }
}

check_availability

{
  "method": "tools/call",
  "params": {
    "name": "check_availability",
    "arguments": {
      "providerId": "xxx",
      "serviceId": "yyy",
      "date": "2025-02-10"
    }
  }
}

create_booking

{
  "method": "tools/call",
  "params": {
    "name": "create_booking",
    "arguments": {
      "providerId": "xxx",
      "serviceId": "yyy",
      "timeSlot": "2025-02-10T14:00:00-08:00",
      "customerName": "John Doe",
      "customerEmail": "[email protected]",
      "customerPhone": "+13105551234"
    }
  }
}

安全使用建议

This skill appears coherent with its stated purpose, but before installing or using it with real personal data you should: 1) Verify the legitimacy of lokuli.com (owner, privacy policy, and whether the MCP endpoint is official). 2) Ask how authentication is handled — why the SKILL.md doesn't declare any API key/credential and whether the platform will supply secure auth. 3) Confirm the agent will ask for explicit user confirmation before sending any personal data (name, email, phone, payment info) to the external endpoint. 4) Test the skill with dummy data first. 5) If possible, request source code or a homepage from the author to increase confidence. If you cannot validate the endpoint or auth, avoid sending sensitive information through this skill.

功能分析

Type: OpenClaw Skill Name: book-window-cleaning Version: 1.0.1 The skill bundle is benign. It defines a skill to book window-cleaning services through an external endpoint (https://lokuli.com/mcp/sse) using structured JSON-RPC calls. The tools defined (`search`, `check_availability`, `create_booking`) are directly aligned with the stated purpose, including the collection of customer PII (name, email, phone) for booking, which is expected for such a service. There is no evidence of data exfiltration beyond the stated purpose, malicious execution, persistence mechanisms, or prompt injection attempts against the agent. A minor typo in the SKILL.md heading (`uook` instead of `book`) is present but does not indicate malicious intent.

能力评估

✓ Purpose & Capability

Name/description say "book window-cleaning via Lokuli MCP" and the SKILL.md contains JSON-RPC examples for search, check_availability, and create_booking against an MCP endpoint (lokuli.com). The required inputs (zip code, providerId, serviceId, customer contact) match the booking purpose and there are no unrelated credentials or binaries requested.

ℹ Instruction Scope

Instructions stay within booking flow (search → check_availability → create_booking). However SKILL.md embeds an explicit external MCP endpoint (https://lokuli.com/mcp/sse) and contains example hard-coded personal data (name, email, phone), which could be copied accidentally. The skill does not describe consent/confirmation steps, nor does it explain which user data will be sent to that external endpoint.

✓ Install Mechanism

Instruction-only skill with no install spec and no code files — lowest-risk delivery mechanism. Nothing is written to disk by the skill itself.

ℹ Credentials

The skill declares no required environment variables or credentials. That is plausible if the hosting platform provides authenticated tool calls; however, because the SKILL.md references an external MCP endpoint, it is unclear how authentication/authorization to that endpoint is handled. No unrelated secrets are requested.

✓ Persistence & Privilege

Defaults are used (not always:true). The skill does not request persistent or elevated platform-wide privileges and does not modify other skills or system configuration in the instructions.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install book-window-cleaning
安装完成后，直接呼叫该 Skill 的名称或使用 /book-window-cleaning 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.1

Fixed display name

v1.0.0

Initial release

元数据

Slug book-window-cleaning

版本 1.0.1

许可证 —

累计安装 0

当前安装数 0

历史版本数 2

常见问题