|
Download README.md from crawlcheck/resolve-snapshot: direct link, hf CLI and curl.
- Browser
- Download file 944 Bytes
-
https://huggingface.co/datasets/crawlcheck/resolve-snapshot/resolve/main/README.md
- Command line
-
hf download hf://datasets/crawlcheck/resolve-snapshot/README.md
-
curl -L -o README.md https://huggingface.co/datasets/crawlcheck/resolve-snapshot/resolve/main/README.md
944 Bytes
| license: cc-by-4.0 | |
| pretty_name: "CrawlCheck Resolve snapshot (daily, signed)" | |
| language: | |
| - en | |
| tags: | |
| - web-measurement | |
| - signed-data | |
| - verifiable | |
| - website-audit | |
| - crawlcheck | |
| # CrawlCheck: Resolve snapshot (daily, signed) | |
| Every public signed answer about a measured site, one file set a day; each line is a ResolveV1 answer that verifies on its own. Newest day only; older days stay on crawlcheck.io. | |
| Mirrored from [crawlcheck.io/spec/resolve](https://crawlcheck.io/spec/resolve) by the CrawlCheck Worker; the site is the canonical copy and is updated first. Measured by CrawlCheck's own observer. Free under CC BY 4.0: credit "CrawlCheck" with the URL of the file. | |
| Each edition is a signed manifest (`manifest.json`, Ed25519 signature inside) and the parts it lists; verify a part by its sha256 in the manifest. | |
| - Catalogue of every dataset: https://crawlcheck.io/datasets | |
| - How the data is made: https://crawlcheck.io/how-it-works | |