About StrictScore

An AI model writes each coin's research inputs: six pillar ratings with rationales, the risk score with factors, and the sources. Each report names its input set by date.

The STRICT engine computes every published number from those inputs and market data. No model writes a score, a probability or a price scenario directly.

• New blog posts are published only after Kamyar Taher, Editor-in-Chief, approves them, and an approved post names him with the line “Reviewed by Kamyar Taher, Editor-in-Chief”. Posts written before October 2026 show that line once he has approved them.

  Coin analyses carry no personal sign-off. No version of the site goes live until every analysis passes an automated check of its data format and of its score ranges.

  When our refresh process updates an analysis that has research inputs on file, the STRICT engine computes its price targets, cycle potential and probability from those inputs and market data, and only after the inputs pass validation: each scored component must give its reasoning and cite at least one source, and no source may be a page of StrictScore itself or only a site’s home page; if the inputs fail, those numbers stay as they were.

  These checks do not mean every source was re-read, and a drafting error or a wrong source can still reach a report.

• There are no author personas and no author pages. Clashware Sàrl, Lausanne, is responsible for the method and the engine.
• Every number is traceable: ratings, weights, arithmetic, numbered sources and the engine stamp with its inputs hash.
• Found an error? Tell us through the contact page, naming the coin and the pillar.

It is research, not financial advice. The engine signal, the 12-month probability and the cycle scenario are model outputs; calibration is provisional.