# Knowledge tools for AI agents: 13 tools, $0.001 to $0.002 per call

Whizbrew has 13 Knowledge tools. One call that returns data costs $0.001 to $0.002; a call that finds nothing or fails costs $0. They are part of one catalogue of 893 tools behind one MCP server and one API key ([all groups](https://whizbrew.com/tools?a=3937d5241035), [pricing](https://whizbrew.com/pricing?a=3937d5241035)).

## Example calls

`wikidata.socials` ($0.001): Official website and every known social account (X, Instagram, TikTok, YouTube, LinkedIn and more) of a brand, company, app or person, with the last recorded follower counts.

```
curl -s -X POST https://whizbrew.com/v1/call/wikidata.socials \
  -H "Authorization: Bearer $WHIZBREW_KEY" -H "content-type: application/json" \
  -d '{"query":"duolingo"}'
```

`wikidata.entity` ($0.001): Facts about a company, brand, app or person: what it is, country, founders, owners, dates, employees, revenue, stock ticker, and the codes other registers know it by (LEI, US regulator number, UK company number, French SIREN, ISIN). fact_ids gives the item ids of its owner, parent and developer, to read that company next.

```
curl -s -X POST https://whizbrew.com/v1/call/wikidata.entity \
  -H "Authorization: Bearer $WHIZBREW_KEY" -H "content-type: application/json" \
  -d '{"query":"shopify"}'
```

`wikipedia.summary` ($0.001): The opening summary of the encyclopedia article about a topic.

```
curl -s -X POST https://whizbrew.com/v1/call/wikipedia.summary \
  -H "Authorization: Bearer $WHIZBREW_KEY" -H "content-type: application/json" \
  -d '{"title":"TikTok"}'
```

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 |
|---|---|---|---|
| `wikidata.socials` | Official website and every known social account (X, Instagram, TikTok, YouTube, LinkedIn and more) of a brand, company, app or person, with the last recorded follower counts. | $0.001 | $1 |
| `wikidata.entity` | Facts about a company, brand, app or person: what it is, country, founders, owners, dates, employees, revenue, stock ticker, and the codes other registers know it by (LEI, US regulator number, UK company number, French SIREN, ISIN). fact_ids gives the item ids of its owner, parent and developer, to read that company next. | $0.001 | $1 |
| `wikipedia.summary` | The opening summary of the encyclopedia article about a topic. | $0.001 | $1 |
| `wikipedia.search` | Search encyclopedia articles by keyword. | $0.001 | $1 |
| `sec.filings_search` | Full-text search of the filings of US-listed companies (annual and quarterly reports, 8-K): which companies wrote about a brand, platform or topic. | $0.001 | $1 |
| `wikipedia.revisions` | The newest edits of an encyclopedia article: time, user, comment, page size and size change, minor flag and a link to the change. A sign of how closely a brand's page is watched. The title must exist (redirects are followed); use wikipedia.search to find it. | $0.001 | $1 |
| `wikipedia.edit_activity` | How many times an encyclopedia article was edited in each of the last months (full months, the running one left out), with the total, the average and the busiest month. Counts edits, not editors. The title must exist (redirects are followed). An article with no edit in the period is free. | $0.002 | $2 |
| `papers.search` | Search scholarly works (journal articles, book chapters, conference papers, preprints) in Crossref by words: title, authors, date, journal or book, publisher, type, how often it is cited, DOI and link. One read of Crossref's public API (metadata publishers register; no full text, no login). In Crossref's order of relevance (its other orders leave the subject, so none is offered). For research-backed content. Nothing found is an empty list. | $0.001 | $1 |
| `papers.work` | One scholarly work by its DOI, from Crossref: title, authors (25 at most), date, journal, volume, issue and pages, publisher, type, how often it is cited, how many references it has, the abstract when the publisher registered one (cut), licence links and full-text links. One read of Crossref's public API; the paper itself is not read, no login. A DOI Crossref does not hold is not_found. | $0.001 | $1 |
| `papers.arxiv_search` | Preprints on arXiv (physics, mathematics, computer science, economics and more) by search words: title, authors, the start of the abstract, the day it was submitted and last updated, categories, and the links to the page and the PDF. Several words are searched as one phrase; a query in arXiv's own field language (ti:, au:, abs:, cat:, joined by AND / OR) is sent as written. Preprints are not all peer reviewed. One read of arXiv's public API, no login, no key, no database of ours. Nothing found is free. | $0.001 | $1 |
| `papers.pubmed_search` | Biomedical and life-science papers in PubMed by search words: title, authors, journal, publication date, publication types, DOI and the link. PubMed's own search syntax works (AND, OR, [Title], [Author]). The list carries no abstracts: they are on each paper's page. Two reads of the US National Library of Medicine's public API (the search, then the summaries), no login, no key, no database of ours. Nothing found is free. | $0.001 | $1 |
| `opendata.eu_search` | Open datasets of European public bodies on data.europa.eu (the EU's portal over the national open-data portals) by search words: title and description in English where the portal has them, publisher, country, catalogue, the days it was issued and changed, file formats and the link. Optional country. Reads the catalogue only, not the data files. One read of the portal's public search, no login, no key, no database of ours. Nothing found is free. | $0.001 | $1 |
| `worldbank.documents_search` | Documents and reports of the World Bank by search words: working papers, country reports, project documents and briefs, with title, date, document type, countries, authors, the start of the abstract and the links to the page and the PDF. For the World Bank's figures by country use market.indicator. One read of the World Bank's public document search, no login, no key, no database of ours. Nothing found is free. | $0.001 | $1 |

Inputs, output shape, health and measured success rate of each tool: [https://whizbrew.com/v1/tools](https://whizbrew.com/v1/tools) and [https://whizbrew.com/status#knowledge](https://whizbrew.com/status#knowledge).

## Start

- **An agent on its own**: read [https://whizbrew.com/agents.md?a=3937d5241035](https://whizbrew.com/agents.md?src=body-text&a=3937d5241035). 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?a=3937d5241035`) gives it a key that works at once with the free credit, with no email and no sign-up form. Its human claims the account later.
- **A chat app** (ChatGPT, Claude, Gemini): add the MCP server `https://whizbrew.com/mcp` as 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?a=3937d5241035](https://whizbrew.com/llms.txt?src=body-text&a=3937d5241035). The agent starts a pairing with one request, its user approves with one click, and the agent holds its own key.
- **Plain HTTP**: create an account at [https://whizbrew.com/app#signup](https://whizbrew.com/app#signup), make a key, then `POST https://whizbrew.com/v1/call/<tool>` with `Authorization: Bearer <key>`.
