A crypto trading intelligence platform whose every published claim is designed to be checked by somebody who does not use it. Most crypto data products ask you to trust a number. This one is built so you do not have to. Every scorecard, benchmark, and signal ships with the evidence it rests on, the method that produced it, and — where the claim is a forecast — a hash committed before the outcome existed to influence it.
Naming. The product rename is in flight and no name has been purchased yet. These docs say “Scienter” rather than baking in a name that is about to change. Code identifiers still use the ghostcopy namespace and are copy-pasteable exactly as written.

The problem this exists to solve

Three failure modes make most crypto analytics unusable for an institution:

Survivorship

The best trader in a pool of 30,000 is usually the luckiest, not the most skilled. Ranking by raw Sharpe guarantees you select noise.

Unfalsifiable records

A screenshot, a cherry-picked window, a leaderboard built by whoever profits from it. Nothing an auditor can reproduce.

Silent absence

“No risks found” and “the risk engine was never installed” render identically in almost every product. They support opposite decisions.

How Scienter answers each

Against survivorship, rankings report a Deflated Sharpe Ratio — a multiple-testing correction applied across the entire scanned population, not just the shortlist that survived. Raw and deflated figures sit next to each other so “best of the population” and “actually skilled” stay visibly different numbers. Against unfalsifiable records, a call’s hash is committed before the outcome exists, and the realized result is bound to that hash exactly once. You can recompute both from the published payload. See Verifiable alpha. Against silent absence, every intelligence response carries an explicit state: ok, not_found, no_snapshot, or module_absent. These are never collapsed. A tool that returns “no risk found” when it means “the risk engine is not installed” is the single worst failure this system can produce, so it is prevented structurally. See Data statesread this before building anything against the API.

What you can actually get today

That last row is the kind of thing most vendors would omit. It is here because a docs site that overstates what is deployed is the first thing an institutional buyer catches, and the last time they trust anything else on the page.

Start here

Quickstart

First API call in five minutes, no key required.

Data states

The four states every response carries, and why collapsing them is a bug.

API surfaces

Three separate HTTP surfaces plus MCP. Which one has what.

Methodology

How each number is produced, including what it cannot show.

Who it is for

See Who it’s for for the four buyer profiles this is built around, and the cases where the honest answer is that you should use something else.
Scienter publishes impersonal market signals of general and regular circulation. Nothing here is investment, legal, or tax advice, or a recommendation to buy, sell, or hold any asset. Scienter is not a registered investment adviser or broker-dealer. Trading digital assets can result in the total loss of your funds — see Disclaimers for the full text.