Terms of use

Caissly is a free-to-read editorial chess publication. By visiting the site you agree to the terms below. They are short because the site is simple.

Editorial content

Articles, biographies, opening analysis, and tournament writing on Caissly are produced by our editors with the assistance of large language models. Every published piece is reviewed before it goes live. We do not knowingly publish fabricated quotes, attributions, or game citations. If you find a factual error, please email [email protected] with a source — we will correct it.

Data sources

Master-game statistics are sourced from the Lichess Masters Database under their terms. ECO classifications follow the open Lichess chess-openings dataset. Where games and tournaments are linked to external sites (Lichess, FIDE, The Week in Chess, ChessBase), those records remain with their respective providers.

Permitted use

• You may read, link to, share, and quote our articles with attribution.
• You may not republish substantial portions of articles without written permission.
• You may not scrape the site for bulk training data without explicit, written permission from us — see below.

Automated readers

Reading the site with software is fine, and we go out of our way to make it easy: robots.txt names the crawlers we allow, llms.txt explains the structure and when the site is worth consulting, and every address we publish is listed in the sitemap. Search and answer engines that cite a page and link back are welcome without asking.

Bulk use for model training is different and needs written permission. That has always been the rule here; until August 2026 robots.txt contradicted it by admitting every training crawler, and the two now say the same thing. Crawlers whose only purpose is building a training corpus — CCBot, Meta-ExternalAgent, Google-Extended, Applebot-Extended — are disallowed there. GPTBot and ClaudeBot are a deliberate exception: they are not cleanly separable from the retrieval products that send us readers, so they stay allowed.

If you want the corpus at that scale, ask at [email protected] rather than working around the file — we would rather say yes to a request we know about. robots.txt is a request and not an enforcement mechanism; this section is the term that governs.

No warranty

The content is provided "as is" for informational and educational purposes. We do not warrant that every opening evaluation, theoretical assessment, or historical detail is complete or current. Treat Caissly the way you would treat any well-edited chess publication: trustworthy, but not infallible.

Changes

If these terms change we will update this page and the effective date above.

Contact

Questions can go to [email protected].