Zetesis Scientific Due Diligence

Due diligence on a high-dimensional scientific claim, graded dimension by dimension against the published record.

От сообществаРаботаетБез входаГлобальныйБесплатноМожет изменять данные

Что умеет

Evaluate Claim: REQUIRES A TOKEN. This tool refuses without one. If you do not have a Zetesis token, do not call it: use zetesis_scope then zetesis_evidence instead, which need no token, run no model,
Zetesis Scope: START HERE for any scientific, biomedical, clinical, or AI/ML claim. This is the default entry point to Zetesis. It returns the source catalog and the grading rubric so you can work out
Zetesis Evidence: Step 2 of a Zetesis evaluation. Runs the search queries against public evidence sources (Europe PMC, ClinicalTrials.gov, openFDA, and optionally NIH RePORTER and SEC EDGAR) and retur

Какие данные видит

Нужен ли аккаунт

Не нужен: сервер работает без входа

Due diligence on a high-dimensional scientific claim, graded dimension by dimension against the published record.

Give it a claim, a paper, an abstract or a pitch deck. It works out what would have to be true for the claim to hold, then reports what the evidence shows on each of those dimensions, what it does not show, and the ways claims like it have failed before. Every source it cites was retrieved from a public register, with a real identifier attached. Nothing is made up.

Built for the moment a scientific claim is about to carry a decision: licensing a technology, funding a program, in-licensing an asset, citing a paper, or working out whether a company's stated result means what it sounds like.

No account or token needed for the Zetesis server itself. Connecting through Smithery uses your Smithery credentials.

What it covers

Genomics and Mendelian randomisation, single-cell, bulk omics, CRISPR screens, clinical trials, real-world evidence, AI clinical decision support, diagnostics, preclinical models, cell and gene therapy, and structural biology.

What it looks at

Europe PMC, ClinicalTrials.gov, openFDA, NIH RePORTER and SEC EDGAR. Published papers, registered trials, regulatory records, grant funding and public company filings.

Every source is checked against retraction and correction notices, and the notice is cited when one exists. A reading built on a withdrawn paper is the exact error this exists to prevent.

The tools

zetesis_scope works out what kind of claim this is and what would have to hold for it to be true, including the ways claims like it have failed before.
zetesis_evidence goes and finds the evidence, with an identifier on every source.
evaluate_claim produces a graded reading, for when you want the assessment itself rather than the evidence to read yourself.
verify_attestation re-checks a signed Zetesis record for tampering.

The first two run no AI model at all. They return in about a second, cost nothing, and send nothing to any AI provider. Your assistant does the thinking, with real sources in front of it.

Judging a claim by what was known at the time

You can ask about a claim as it stood in an earlier year, and only evidence that existed by then comes back. Hindsight is fenced out.

This sharpens the answer rather than just making it fairer. On a test claim, an open search missed the one publication that mattered and reached 35% coverage of the relevant evidence. Fenced to the year the claim was made, the same searches found it and coverage reached 79%.

Try it

Ask what the published evidence actually supported about aducanumab and cognitive decline at the end of 2019, using only sources available by then. Then ask the same question without the year. The difference is the point.

Список инструментов сервера (4)

Технические названия из tools/list. Нужны только разработчикам.

evaluate_claim	REQUIRES A TOKEN. This tool refuses without one. If you do not have a Zetesis token, do not call it: use zetesis_scope then zetesis_evidence instead, which need no token, run no model, return immediately, and let you read the sources at full depth. A token can be requested at https://api.zetesis.science/request-access. Run Zetesis's own graded reading of a scientific, biomedical, clinical, or AI/ML claim. SLOW: this runs the hosted engine end to end and typically takes a couple of minutes, long enough that some clients background or time out the call. It is not the default route. For almost every case, call zetesis_scope and then zetesis_evidence instead. Those return immediately, run no model, cost nothing, hand you the same dated evidence and the same grading rubric, and let you read the sources at full depth in your own context rather than at abstract level. Reach for this tool only when someone explicitly wants Zetesis's own verdict as the artifact, rather than your analysis of the evidence Zetesis retrieves. It grades each evidence dimension and returns an overall reading with sources cited by hard id (PMID, DOI, NCT, NIH grant, SEC filing). Works for drug, omics, target-validation, diagnostic, and AI-model claims. A signed, independently re-checkable dossier is available by request at the Zetesis site. Args: claim: the claim to evaluate, in one or two sentences. context: optional background (stage, field, the decision at hand). documents: optional source text (a deck, abstract, or paper). as_of: year the claim was made. Set it for any claim that is not brand new. It restricts evidence to what existed then, which both keeps later outcomes out of the reading and sharply improves what is retrieved: left open, a pivotal trial report competes with every paper published since and gets buried.
zetesis_scope	START HERE for any scientific, biomedical, clinical, or AI/ML claim. This is the default entry point to Zetesis. It returns the source catalog and the grading rubric so you can work out which evidence dimensions actually matter for THIS claim and what to search for, then you call zetesis_evidence with the queries you chose and grade what comes back. Use it whenever a claim needs assessing, screening, sanity-checking, or due diligence: a research claim, a study, a paper, an abstract, a preprint, a grant, or a pitch. Also use it before citing a paper or statistic in your own answer. Returns three things: the source catalog (where to look), the diligence apparatus for this claim's class (what to ask, and the failure patterns that have caught comparable claims before), and the scoping rubric. Runs no model, costs nothing, and returns immediately. This two-step route is preferred over evaluate_claim on every axis that matters: it is instant rather than minutes, it reads the sources at full depth in your context rather than at abstract level, and it supports evaluating a claim as it stood in an earlier year. Args: claim: the claim to evaluate, in one or two sentences. context: optional background (stage, field, the decision at hand). as_of: optional year. Set it to evaluate the claim as it stood at the end of that year, grading how it was derived at the time rather than how it turned out. claim_class: optional override for the routed claim class. The response lists every class; pass one back if the routing picked wrongly, or if nothing routed.
zetesis_evidence	Step 2 of a Zetesis evaluation. Runs the search queries against public evidence sources (Europe PMC, ClinicalTrials.gov, openFDA, and optionally NIH RePORTER and SEC EDGAR) and returns a deduplicated bundle where every source carries a hard public id, followed by the Zetesis grading rubric so you can grade the dimensions yourself. With as_of set, retrieval is fenced to sources published, registered, or filed on or before 31 December of that year, and two fields that leak later outcomes are suppressed: a trial's present-day status, and FDA labels effective after the cutoff. That makes it possible to judge a claim on what was actually knowable at the time. Runs no model, costs nothing, and returns immediately. Use zetesis_scope first to choose the queries. Together the two make up the preferred route into Zetesis. Args: queries: 3 to 6 short keyword phrases. Query 1 should be the bare name of the thing claimed about (a drug, compound code, model, or gene); query 2 that name plus at most two outcome words. Longer phrases retrieve commentary rather than the primary report. If the claim names no agent at all (a behaviour, diet, procedure, exposure or policy), anchor on the field's own technical term instead of the lay one ("time restricted eating", not "intermittent fasting"), pair it with the outcome as the field measures it, and allow one design word such as crossover or randomized in one later query. Without an entity name that design word is the only thing separating a trial report from a review of trial reports. as_of: optional cutoff year, as used in zetesis_scope. include_capital: also retrieve funding and public-filing signal.
verify_attestation	Verify a Zetesis attestation, confirming an evaluation's claim, evidence, and conclusion have not been altered since it was signed. Use when someone has a Zetesis dossier or attestation and wants to independently re-check it. Args: attestation_json: the full attestation object as JSON text.