Reviews tools for AI agents: 11 tools, $0.000001 to $0.000136 per call
Whizbrew has 11 Reviews tools. One call that returns data costs $0.000001 to $0.000136; a call that finds nothing or fails costs $0. They are part of one catalogue of 920 tools behind one MCP server and one API key (all groups, pricing).
Example calls
trustpilot.summary ($0.000002): Trustpilot score of a company: TrustScore, stars, review total and the count per star. Needs the company's Trustpilot business unit id, which you must supply: it cannot be looked up from a domain or name here.
curl -s -X POST https://whizbrew.com/v1/call/trustpilot.summary \
-H "Authorization: Bearer $WHIZBREW_KEY" -H "content-type: application/json" \
-d '{"business_unit_id":"46d6a890000064000500e0c3"}'
trustpilot.reviews ($0.000009; with limit = 1 to 10: $0.000002): Newest Trustpilot reviews of a company (stars, title, text, author, date). Needs the company's Trustpilot business unit id, which you must supply: it cannot be looked up from a domain or name here. Long texts can be cut (text_is_cut).
curl -s -X POST https://whizbrew.com/v1/call/trustpilot.reviews \
-H "Authorization: Bearer $WHIZBREW_KEY" -H "content-type: application/json" \
-d '{"business_unit_id":"46d6a890000064000500e0c3","limit":3}'
ratings.film ($0.000022): Rating of a film or TV show from Letterboxd (member average out of 5) or Rotten Tomatoes (Tomatometer and audience score out of 100).
curl -s -X POST https://whizbrew.com/v1/call/ratings.film \
-H "Authorization: Bearer $WHIZBREW_KEY" -H "content-type: application/json" \
-d '{"slug":"parasite-2019"}'
Over MCP the same call is call_tool with {"tool": "<name>", "arguments": {…}}.
Every tool in this group
| Tool | What it returns | Price per call that returns data | Per 1,000 calls |
|---|---|---|---|
trustpilot.summary | Trustpilot score of a company: TrustScore, stars, review total and the count per star. Needs the company's Trustpilot business unit id, which you must supply: it cannot be looked up from a domain or name here. | $0.000002 | $0.002 |
trustpilot.reviews | Newest Trustpilot reviews of a company (stars, title, text, author, date). Needs the company's Trustpilot business unit id, which you must supply: it cannot be looked up from a domain or name here. Long texts can be cut (text_is_cut). | $0.000009; with limit = 1 to 10: $0.000002 | $0.009 |
ratings.film | Rating of a film or TV show from Letterboxd (member average out of 5) or Rotten Tomatoes (Tomatometer and audience score out of 100). | $0.000022 | $0.022 |
goodreads.search | Top five Goodreads books for a title or author, with average rating, rating count and pages. | $0.000004 | $0.004 |
books.search | Search books in the Open Library catalogue: title, authors, first year, editions, rating and ISBN. | $0.000064; with limit = 1: $0.000001; 2 to 3: $0.000003; 4 to 10: $0.000004 | $0.064 |
smartcustomer.reviews | Customer reviews of a business by its website domain, from SmartCustomer (the review site that was Sitejabber): the business name, rating out of 5 and review count, and 20 reviews a page, newest first (stars, title, text, reviewer name, date, link). "page" reads older reviews; has_more says whether another page exists. Covers businesses with a website that customers reviewed there (mostly online shops and services, US-heavy); a site without a review page answers not found. | $0.000028 | $0.028 |
reviews.for_business | One business across every open review source in one answer, with the source of each number: Google Maps (rating, review count, count per star, themes, newest reviews), the Bing Maps and Apple Maps listings (the rating they show and the review site it comes from, e.g. Tripadvisor or Yelp), SmartCustomer (by the website domain), the business's own website (the rating it states in its structured data) and Trustpilot (only when trustpilot_id is given). Give the name and the city or address; "place_id" or "domain" make the match exact. Each source is a line with its status (found, listed_without_rating, not_found, skipped, failed) and the tool that read it; ratings are not averaged. For more Google reviews use maps.reviews with the place_id of the answer. | $0.000136 | $0.136 |
goodreads.list | The books of a public Goodreads list (e.g. Best Books Ever), 100 a page in the list's order: rank, title, author, average rating, ratings count, list score and votes. Pages after the first can take several seconds. | $0.000049 | $0.049 |
books.subject | Books filed under a subject in the Open Library catalogue (e.g. marketing, science fiction): title, authors, first year, number of editions and cover. A subject with no books is free. Can take a few seconds. | $0.000016; with limit = 1 to 3: $0.000002; 4 to 5: $0.000003; 6 to 20: $0.000005 | $0.016 |
books.ratings | The reader ratings of one book on Open Library: average, count and the count per star. Takes the work id that books.search and books.subject answer with. Far fewer ratings than Goodreads. A work nobody rated (or an unknown id: the source does not tell them apart) is free. Can take a few seconds. | $0.000001 | $0.001 |
reviews.summary | What reviewers say, without a language model: reads the newest reviews of one app, product, shop, plugin, game or podcast and computes the average rating, the spread over 1 to 5 stars, the share of positive reviews, the ratings month by month, whether the newer half is better or worse than the older half, and the words and two-word phrases found in the most reviews: over all of them, in the 1 and 2 star reviews (what people complain about) and in the 4 and 5 star reviews (what they praise), with example quotes. Sources: appstore, playstore, trustpilot, steam, shopify_app, wordpress_plugin, firefox_addon, microsoft_store, woocommerce, lazada, podcast. Computed from one read of the newest reviews, not from all reviews; the source's own totals are passed along where it gives them. For a Google Maps place use maps.reviews_summary. | $0.000019; with limit = 5: $0.000002 | $0.019 |
Inputs, output shape, health and measured success rate of each tool: https://whizbrew.com/v1/tools and https://whizbrew.com/status#reviews.
Start
- An agent on its own: read https://whizbrew.com/agents.md. It asks for a free quote (
GET https://whizbrew.com/v1/agent/guide?goal=<the task in plain words>), then one request (POST https://whizbrew.com/v1/agent/start) gives it a key, with no sign-up form; data calls start once its human confirms an email with a code (one question), which brings the free credit. - A chat app (ChatGPT, Claude, Gemini): add the MCP server
https://whizbrew.com/mcpas a custom connector and sign in. No key to copy. - An agent with a shell (Claude Code, Codex, Cursor): read https://whizbrew.com/llms.txt. The agent starts a pairing with one request, its user approves with one click, and the agent holds its own key.
- Plain HTTP: an agent or a script gets a key with
POST https://whizbrew.com/v1/agent/start(no email, no sign-up form); a person creates an account at https://whizbrew.com/app#signup and makes a key there. ThenPOST https://whizbrew.com/v1/call/<tool>withAuthorization: Bearer <key>.