← 返回 Skills 市场
cnnrobrn

Apartment Cleaning

作者 cnnrobrn · GitHub ↗ · v1.0.7 · MIT-0
cross-platform ✓ 安全检测通过
133
总下载
1
收藏
0
当前安装
8
版本数
在 OpenClaw 中安装
/install claw-cleaning
功能描述
Book a professional apartment cleaning in San Francisco via the claw.cleaning REST API. Use when someone wants to book, schedule, or inquire about apartment...
使用说明 (SKILL.md)

Apartment Cleaning Booking

At a Glance

  • Service: Professional apartment cleaning
  • Area: San Francisco only
  • Rate: $40/hour (1–8 hours)
  • Days: Any day of the week on/after 2026-05-05. Before then, Saturdays and Sundays only.
  • Hours: 8 AM – 6 PM PT
  • Payment: No upfront payment. The customer pays the cleaner (cash or card) at the appointment.

How This Skill Works

This skill calls the claw.cleaning REST API directly over HTTPS — no local CLI, no MCP server.

  • Base URL: https://claw.cleaning
  • Auth: None. All endpoints are public.
  • Content type: application/json

Use the agent's HTTP/fetch capability (e.g. curl, fetch, requests) to call the endpoints below.

Available Endpoints

  • GET /availability — list open slots
  • POST /bookings/initiate — reserve a slot (calendar invite sent immediately, customer pays the cleaner at the appointment)
  • GET /bookings/status — list upcoming bookings by email

Safety Rules

  • Never POST /bookings/initiate without showing the full preview to the customer and getting explicit confirmation ("yes", "confirm", "book it", etc.).
  • Always GET /availability before POST /bookings/initiate to confirm the slot is listed as available.
  • Do not invent available times — only offer times returned by GET /availability.
  • Make it clear to the customer that the total ($40/hour × hours) is paid in cash or card to the cleaner at the end of the session.

Booking Workflow

Step 1 — Check availability

GET https://claw.cleaning/availability returns the next 14 upcoming bookable days under a top-level days array. Pass ?date=YYYY-MM-DD to check one specific day. Before 2026-05-05 the date must be a Saturday or Sunday; on or after, any day of the week is bookable.

Examples:

GET https://claw.cleaning/availability
GET https://claw.cleaning/availability?date=2026-05-06

Present the available slots clearly. Each slot shows startTime and maxHours available.

Step 2 — Collect customer details

Ask the customer:

  • Which date and start time?
  • How many hours? (1–8, $40 each)
  • Full SF address (street, city, state)
  • Their name
  • Their email (calendar invite goes here)

Step 3 — Confirm before booking

Show the customer a summary (date, start time, hours, address, total, email) and remind them the total is paid to the cleaner at the appointment. Ask for explicit confirmation before posting the booking.

Step 4 — Initiate booking

POST https://claw.cleaning/bookings/initiate with a JSON body:

{
  "date": "YYYY-MM-DD",
  "startTime": "HH:MM",
  "hours": 1-8,
  "address": "... San Francisco, CA ...",
  "name": "Customer Name",
  "email": "[email protected]"
}

Send Content-Type: application/json. On success returns { "status": "booked", "total": "$X", ... }. The slot is reserved immediately and the calendar invite is sent.

Step 5 — Deliver the outcome

Tell the customer the slot is booked, the calendar invite is on its way, and they owe the cleaner $40/hour at the end of the session (cash or card).

Step 6 — Check status (optional)

GET https://claw.cleaning/bookings/status?email=\x3Curl-encoded-email> if the customer asks whether their booking went through. Returns upcoming bookings for that email.

Key Details

  • Working hours: 8 AM – 6 PM PT
  • 30-min travel buffers are automatically blocked before and after each cleaning
  • The calendar event blocks the slot immediately on booking
  • Persistent no-shows may result in the email being blocked from future bookings

Error Handling

All errors return a JSON body with an error field and a non-2xx HTTP status.

  • 400 "Address must be in San Francisco, CA." → ask for a valid SF address
  • 400 "Hours must be between 1 and 8." → correct the hours value
  • 400 "Weekdays before 2026-05-05 are not bookable..." → pick a Saturday/Sunday, or any day on or after 2026-05-05
  • 400 "Cannot book a time in the past..." → pick a future date and time
  • 400 "Invalid date format. Use YYYY-MM-DD." / "Invalid startTime format. Use HH:MM (24h)." → fix the format
  • 400 "Invalid email address." → ask for a valid email
  • 403 "This email is blocked from booking." (code: "email_blocked") → the operator has blocked this email due to prior no-shows. Customer should contact [email protected].
  • 409 "That time slot is no longer available." → call GET /availability and offer alternatives
  • 500 "Failed to fetch availability." / "Failed to initiate booking." / "Could not create calendar event." → retry once; if it keeps failing, tell the customer to try again later

See references/booking-flow.md for a full example conversation flow with sample HTTP requests and responses.

安全使用建议
This skill appears coherent, but you should still verify the service domain (https://claw.cleaning) is trustworthy before sharing personal data or making real bookings. Ensure the agent shows the full booking preview and you explicitly confirm before it sends a POST. Be aware the service claims no upfront payment (customer pays the cleaner in person), and avoid automating large numbers of bookings without rate-limit/backoff logic. If you need stronger privacy, use an email you control for test bookings and confirm cancellation/contact details ([email protected]) before relying on it.
功能分析
Type: OpenClaw Skill Name: claw-cleaning Version: 1.0.7 The skill is a straightforward integration for booking apartment cleaning services in San Francisco via a dedicated REST API (claw.cleaning). It follows a standard, transparent workflow that includes checking availability, collecting necessary user details (name, email, address), and requiring explicit user confirmation before initiating a booking. There are no signs of malicious intent, data exfiltration, or unauthorized command execution.
能力标签
cryptocan-make-purchases
能力评估
Purpose & Capability
Name/description match the behavior: the SKILL.md exclusively describes checking availability, collecting customer details, and posting bookings to https://claw.cleaning. No unrelated services, binaries, or credentials are requested.
Instruction Scope
Runtime instructions are narrowly scoped to making HTTPS requests (GET/POST) to the documented endpoints, collecting user-provided booking details, and showing a confirmation before booking. The instructions do not ask the agent to read local files, environment variables, or other system state.
Install Mechanism
No install spec or code is included (instruction-only), so nothing will be written to disk or downloaded during install.
Credentials
The skill requires no environment variables, credentials, or config paths. It only uses customer-supplied data (name, email, address) appropriate for booking.
Persistence & Privilege
always is false and the skill is user-invocable; it does not request permanent presence or to modify other skills or system settings.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install claw-cleaning
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /claw-cleaning 触发
  4. 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
v1.0.7
Apartment cleaning can now be booked any day of the week on/after 2026-05-05. - Expanded booking to allow any day of the week for dates on/after 2026-05-05 (previously weekends only). - Availability endpoint now returns the next 14 upcoming bookable days and includes both weekends and weekdays if applicable. - Updated error handling messages for non-bookable weekdays and past times. - Clarified instructions and descriptions to reflect new booking rules while maintaining all prior workflows and safety checks.
v1.0.6
claw-cleaning 1.0.6 changelog: - Switched integration from the MCP server to a direct REST API over HTTPS for all booking and availability actions. - Updated connection instructions: now uses public, unauthenticated REST endpoints (`/availability`, `/bookings/initiate`, `/bookings/status`). - Clarified error handling: added HTTP status codes, API error messages, and JSON request/response formats. - Improved workflow steps and endpoint usage examples for clarity and precision. - Removed references to streamable HTTP and MCP connectors; updated technical notes throughout.
v1.0.5
**Major update: Booking flow now requires payment only at the appointment (no upfront payment), and support for card payments in advance has been removed.** - Eliminated upfront/Stripe payment: customers now pay the cleaner directly (cash or card) at the appointment. - Removed all references to "pay now" options, Stripe, and customer credentials from flow and documentation. - Booking workflow and available tools simplified to reflect a pay-at-appointment only process. - Adjusted safety rules and error handling for the new payment expectations. - Only these API tools are now supported: `check_availability`, `initiate_booking`, and `check_booking_status`.
v1.0.4
- Added support for gated Stripe repeat-customer auto-charges using a username and password pair, minted after first successful checkout. - Introduced the `claim_credentials` tool to fetch and store customer credentials for secure repeat charges. - Updated booking flow: returning Stripe customers must provide stored credentials for auto-charging; if missing or incorrect, a new Stripe checkout is triggered to re-verify and re-mint credentials. - Clarified credential usage rules: never display or recite credentials to the customer, and only use them for pay-now repeat Stripe bookings. - Booking and payment instructions updated to reflect new credential and security requirements.
v1.0.3
- Improved clarity and organization of instructions in SKILL.md - Added an "At a Glance" summary for key service details (area, rate, hours, payment) - Grouped endpoint and tool information for easier setup and troubleshooting - Made workflow steps and safety rules more scannable - No functional or behavioral changes to skill logic
v1.0.2
Apartment-cleaning skill now uses the MCP server instead of the CLI. - Switched backend from local CLI to the `claw-cleaning` MCP server at https://claw.cleaning/mcp. - Updated instructions and safety rules to reference MCP tools: `check_availability`, `initiate_booking`, `force_checkout_booking`, `check_booking_status`. - No installation needed; users now connect via MCP-compatible clients. - Clarified booking workflow and error handling to match new API responses. - Payment and confirmation flows updated for MCP system and new response formats.
v1.0.1
claw-cleaning v1.0.1 adds pay-on-completion booking and switches to the new CLI. - Switched CLI from clawt to claw-cleaning; metadata and commands updated accordingly - No configuration needed for end users; CLI points to production automatically - Added support for two payment options: pay now (Stripe), or pay at appointment (cash/card) - Booking workflow and confirmation flow updated to handle both payment types - Explicit safety rules for payment choice, confirmation, and link sharing
v1.0.0
- Initial release of the apartment-cleaning booking skill. - Supports booking, scheduling, and inquiries for apartment cleaning in San Francisco (weekends only, $40/hour). - Guides users through the full workflow: availability check, collecting customer details, confirmation, booking, payment, and status check. - Enforces safety and booking rules, including confirmation before booking and offering only available time slots. - Provides error handling for address, time slot, and hours validation. - Adds zero-install alternative using Model Context Protocol (MCP) for compatible clients.
元数据
Slug claw-cleaning
版本 1.0.7
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 8
常见问题

Apartment Cleaning 是什么?

Book a professional apartment cleaning in San Francisco via the claw.cleaning REST API. Use when someone wants to book, schedule, or inquire about apartment... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 133 次。

如何安装 Apartment Cleaning?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install claw-cleaning」即可一键安装,无需额外配置。

Apartment Cleaning 是免费的吗?

是的,Apartment Cleaning 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Apartment Cleaning 支持哪些平台?

Apartment Cleaning 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Apartment Cleaning?

由 cnnrobrn(@cnnrobrn)开发并维护,当前版本 v1.0.7。

💬 留言讨论