UncleCode

license

Crawl4AI-Attribution-License(Apache-2.0 text plus an appended attribution condition covering public use

source

public(the published repository builds the crawler, the Docker deployment and the CLI)

core features withheld

no — a Cloud API is announced as a separate closed beta beside the library, not carved out of it

Crawl4AI's LICENSE carries the standard Apache-2.0 text, but appends an Attribution Requirement covering any public use, not just redistribution -- a credit line in a NOTICE, a publication's acknowledgments, a site's About page or a CLI's help output. Because Apache-2.0 only binds redistribution, this condition falls outside what the OSI approved, even though GitHub's classifier and badge still read Apache-2.0. The repository is otherwise public and ungated, with no closed core behind it.

• https://api.github.com/repos/unclecode/crawl4ai recorded 2026-08-30

  Repo metadata reports license spdx_id Apache-2.0 and 80,161 stars; the API does not detect the appended attribution section that the LICENSE file itself carries.

• https://raw.githubusercontent.com/unclecode/crawl4ai/main/LICENSE recorded 2026-08-30

  LICENSE body: the stock Apache-2.0 text, then after 'END OF TERMS AND CONDITIONS' a separate 'Attribution Requirement' section - 'All distributions, publications, or public uses of this software, or derivative works based on this software, must include the following attribution', with placement required in a NOTICE/README, a publication's acknowledgments, a website's About or Credits section, or a command-line tool's help output.

• https://raw.githubusercontent.com/unclecode/crawl4ai/main/README.md recorded 2026-08-30

  README documents the crawler, Docker deployment and CLI built from this repository, and announces a Crawl4AI Cloud API as a separate closed beta. No enterprise directory, license key or paid build of the library itself is described.

Adoption is measured by PyPI downloads of the crawl4ai package over the trailing month, which is the closest first-party channel to how the library is used.

• https://pypistats.org/api/packages/crawl4ai/recent recorded 2026-08-30

  last_month = 1,524,522 downloads for crawl4ai.

Crawl4AI's core interface, `arun`, takes a URL to fetch rather than a query to search, and its crawling follows links from that URL rather than searching an index. Because it never accepts a query, it covers only the retrieval half of a find-then-extract workflow, not the discovery half a search interface would add.

• https://raw.githubusercontent.com/unclecode/crawl4ai/main/README.md recorded 2026-09-17

  The quickstart: `async with AsyncWebCrawler() as crawler: result = await crawler.arun(url="https://www.nbcnews.com/business")`, and the CLI `crwl https://www.nbcnews.com/business -o markdown`. Both take a URL. No query-based search interface appears in the README.