# Ground an answer or enrich records from Wikipedia and Wikidata

> Ground an answer or enrich records from Wikipedia and Wikidata: 1 companies fully do it, 0 with a limit, 3 carry the tag. Every listing in Data providers with the reason and a source.

- Canonical: https://www.anchorterminal.com/niches/data-providers.knowledge-grounding/
- Markdown: https://www.anchorterminal.com/niches/data-providers.knowledge-grounding/index.md (~1,100 tokens)
- Slim: https://www.anchorterminal.com/niches/data-providers.knowledge-grounding/index.min.md (~230 tokens, same facts, less prose, for token-sensitive contexts)
- JSON: https://www.anchorterminal.com/niches/data-providers.knowledge-grounding/index.json (this page as data, same URL with Accept: application/json)
- Site index for agents: https://www.anchorterminal.com/llms.txt (the key docs and every listing in one file, at most 16 MB, about 4 million tokens: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- MCP: https://www.anchorterminal.com/mcp (the whole catalogue as tools, no key)
- DNS: dig +short TXT <slug>.grade.anchorterminal.com (the grade as a DNS TXT record, for a sandbox with no HTTP where names still resolve; also top., best.<category>. and <a>.vs.<b>. under the zone, and the same under grade.letme.dev)
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)
- Updated: 2026-11-15

One company's docs show a full fit.

| Companies, full fit | With a limit | Carry the tag | Share of the category's work | First-hand accounts |
| --- | --- | --- | --- | --- |
| 1 | 0 | 3 | 9 | 5 |

## What counts as done

The article text or entity facts with the page URL and revision id, retrieved under the identified-client rate limit, with article text treated as untrusted.

How agents ask for it: what does Wikipedia say about this; get the article and cite it; fill in author and book metadata; link these places to Wikidata

## Does the job

The docs show it meets the test above, and nothing we read says otherwise.

- **Wikimedia REST API (Wikimedia Foundation, Inc.)**: Search pages, then the page object gives the latest revision id, html_url and licence, with page history and revisions by id; a User-Agent with contact raises the limit from 10 to 200 requests a minute. acc-44 warns identifying can expose the user's email.

## Couldn't check

The docs were closed to us, or silent. These carry no mark either way.

- **CompanyLens (CompanyLens)**: companylens.io answered robots.txt with a Cloudflare challenge and was treated as closed, so no docs were read.
- **SEC EDGAR APIs (U.S. Securities and Exchange Commission)**: www.sec.gov answered robots.txt with a 'Request Rate Threshold Exceeded' page asking automated access to comply with its policy; the refusal was recorded and the docs were not read.

- not this job: Assymetrix MCP Server (Assymetrix Technology, LLC): Prediction-market data only (Polymarket, Kalshi, Limitless); this use case is outside its coverage.
- not this job: Codex (Codex Data Inc.): On-chain crypto and prediction-market data only; this use case is outside its coverage.
- not this job: CoinDesk Data API (CC Data Limited, trading as CoinDesk Data (England and Wales, 10966788)): Crypto market, index and news data; this use case is outside its coverage.
- not this job: CoinGecko x402 API (Gecko Labs Pte. Ltd.): Crypto prices and on-chain pool data only; this use case is outside its coverage.
- not this job: Companies House API (Companies House, an executive agency of the Department for Business and Trade): UK company register only; this use case is outside its coverage.
- not this job: CryptoQuant MCP server (Team CQ Inc.): Crypto market and on-chain data only; this use case is outside its coverage.
- not this job: FRED API (Federal Reserve Bank of St. Louis): Economic time series only; this use case is outside its coverage.
- not this job: Getty Images MCP Server (Getty Images Holdings, Inc.): Licensed stock and editorial images and video only; this use case is outside its coverage.
- not this job: Open-Meteo (OpenMeteo GmbH): Weather data only; this use case is outside its coverage.
- not this job: OpenBB (OpenBB Inc.): No provider in the V5 provider table covers this use case.
- not this job: registercheck.de API (Openlaw GmbH): German commercial register only; this use case is outside its coverage.

## What stops agents finishing

- unidentified clients held to 10 requests a minute
- an agent that identifies itself may leak the user's email

## Around it

Before: search.ground-fact. After: search.verify-citation, company-knowledge.cited-answer.

## Where this comes from

Weight 9 of 100 in Data providers: 5 of 46 accounts (Hacker News: Wikidata enrichment, agent load on Wikimedia, User-Agent identification)

As data: https://www.anchorterminal.com/api/v1/use-cases/data-providers.knowledge-grounding.json, or the MCP tool get_use_case with id data-providers.knowledge-grounding. Draft, read 2026-11-15 from public docs and first-hand accounts, not yet reviewed by a person.

---

Query this catalogue directly over MCP: https://www.anchorterminal.com/mcp (Streamable HTTP, no key). How: https://www.anchorterminal.com/agents/#mcp
