Surface the angriest reviews nobody has answered yet, ranked so your team replies in the order that protects your rating most
https://apidirect.io/mcp?token=YOUR_API_KEY
Build me a triage queue of the angriest unanswered reviews for {brand_name} in {location}, including our Facebook page {facebook_page_url}
one-star-review-triage-queue.
Any agent can also call get_skill(skill_id="one-star-review-triage-queue") to pull these steps on demand.
Negative reviews with no owner response do the most lasting damage, yet they hide at the bottom of the pile. This skill pulls the lowest-ranked, angriest, still-unanswered reviews from both Google Maps and Facebook into one prioritized response queue.
Who it's for: Reputation and CX managers for local or multi-location brands
| Input | Required | Description | Example |
|---|---|---|---|
brand_name |
Yes | The business name to monitor reviews for | Bluebird Coffee |
location |
Yes | City/area to disambiguate the Google Maps listing | Austin, Texas |
facebook_page_url |
No | The brand's Facebook page URL to also pull recommendations from | https://www.facebook.com/bluebirdcoffee |
search_places(query="{brand_name} {location}")
Match the brand and grab the place_id of the correct listing (verify by name and business_status).
place_reviews(place_id=<place_id>, sort_by=lowest_ranking, get_sentiment=true, pages=4)
Keep 1-2 star reviews with no owner response whose dominant_emotion is anger or disgust, hardest-hitting first.
facebook_page_details(url={facebook_page_url})
Resolve the page_id needed to read the page's recommendations.
facebook_page_reviews(page_id=<page_id>, get_sentiment=true, pages=3)
Keep recommend=false items with negative sentiment and no page reply, then interleave with the Google queue by severity.
This is exactly what the MCP returns to your agent (via the one-star-review-triage-queue prompt or get_skill tool), with your inputs filled in.
SKILL: One-Star Review Triage Queue
Negative reviews with no owner response do the most lasting damage, yet they hide at the bottom of the pile. This skill pulls the lowest-ranked, angriest, still-unanswered reviews from both Google Maps and Facebook into one prioritized response queue.
You are running this skill on API Direct via its MCP tools. Execute the steps below yourself by calling the named tools in order — values in <angle brackets> come from a previous step. Then deliver the result described at the end.
INPUTS:
- brand_name (required): <brand_name — ASK THE USER>
- location (required): <location — ASK THE USER>
- facebook_page_url (optional): (optional — e.g. https://www.facebook.com/bluebirdcoffee)
STEPS:
1. Tool `search_places` — search_places(query="<brand_name> <location>")
Match the brand and grab the place_id of the correct listing (verify by name and business_status).
2. Tool `place_reviews` — place_reviews(place_id=<place_id>, sort_by=lowest_ranking, get_sentiment=true, pages=4)
Keep 1-2 star reviews with no owner response whose dominant_emotion is anger or disgust, hardest-hitting first.
3. Tool `facebook_page_details` — facebook_page_details(url=<facebook_page_url>)
Resolve the page_id needed to read the page's recommendations.
4. Tool `facebook_page_reviews` — facebook_page_reviews(page_id=<page_id>, get_sentiment=true, pages=3)
Keep recommend=false items with negative sentiment and no page reply, then interleave with the Google queue by severity.
DELIVER: A single prioritized triage queue of the angriest unanswered Google and Facebook reviews, ordered by sentiment intensity for the fastest rating-saving replies
Note: each underlying tool call is billed at its normal endpoint price; get_sentiment adds a small per-page surcharge. Page through results as needed but stop once you have enough to deliver the outcome.
See — and source-trace — exactly what Google's AI tells the public about your brand.
For a damaging tweet, map who's spreading it, their reach, and the mood of the commentary.
Measure whether your crisis statement actually calmed the room or poured gas on it