This server provides MCP tools for scientific due diligence: evaluating claims against literature, trials, and filings; checking deep evaluations; verifying signed attestations; and checking account status.
evaluate_claim: Assess a scientific, biomedical, clinical, or AI/ML claim with optional context and documents. Use depth="screen" for a fast, free read (no token needed) or depth="deep" for a full dossier with independent re-checking. mode="diligence" adds capital/financial context. Returns graded evidence with hard identifiers (PMID, DOI, NCT, NIH grant, SEC filing).
check_evaluation: Retrieve the result of a previously started deep evaluation using its job_id.
verify_attestation: Verify a signed Zetesis attestation to confirm the claim, evidence, and conclusion have not been altered since signing. No account needed.
account_status: Show the caller's Zetesis plan and remaining free evaluations or dossier credits.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@zetesis evaluate the claim that mRNA vaccines alter DNA"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Scientific due diligence on a claim, from inside Claude, Copilot, or any MCP host.
Give Zetesis a claim, an abstract, a paper, a grant or a deck. It routes the claim to its scientific class, then returns the questions a domain reviewer would ask, the failure patterns that caught comparable claims before, and the public evidence bearing on it, with a PMID, DOI, NCT number, NIH grant number or SEC filing reference on every source. Every identifier it hands back was retrieved. None are generated.
It can also evaluate a claim as it stood in an earlier year, restricting evidence to what existed by then, so a claim is judged on what was knowable at the time rather than on how it turned out.
The hosted server is at https://api.zetesis.science/mcp, over Streamable HTTP.
No account, key or token is required.
Claude Code:
claude mcp add --transport http zetesis https://api.zetesis.science/mcpClaude Desktop (claude_desktop_config.json):
{
"mcpServers": {
"zetesis": {
"type": "http",
"url": "https://api.zetesis.science/mcp"
}
}
}Any other MCP client:
Client | How |
Microsoft Copilot Studio | Tools, then Add a tool, then Model Context Protocol. Server URL, auth None. |
ChatGPT | Settings, then Connectors, then Developer mode. Add the URL. |
Gemini CLI |
|
For Gemini's settings.json, use httpUrl rather than url; the latter is SSE and will not
connect. Full setup notes: https://api.zetesis.science/docs
Related MCP server: traceable-research-mcp
zetesis_scope routes the claim and returns the diligence apparatus for its class: the
questions a reviewer would ask, structured by substrate, methods, cohort and risk of bias, a
failure-pattern taxonomy carrying the companies each pattern was derived from, and the edge cases
where those patterns were wrong. A checklist that only ever fires positive teaches over-rejection,
so the counterexamples ship alongside it.
zetesis_evidence runs the searches and returns a deduplicated bundle from Europe PMC,
ClinicalTrials.gov, openFDA, NIH RePORTER and SEC EDGAR, every source carrying a hard public
identifier, followed by the grading rubric so you grade the evidence yourself in context.
Neither of those calls a language model. They return in under a second, cost nothing to run, and send nothing to a model provider. That is usually the answer a security reviewer is looking for.
evaluate_claim produces Zetesis's own graded reading server-side. Slower, and only needed
when the assessment itself is the deliverable rather than the evidence.
verify_attestation re-checks a signed Zetesis record to confirm its claim, evidence and
conclusion have not been altered since signing. Needs no account.
Claim classes: 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, structural biology.
Ask a general model about a 2020 claim today and it answers with years of hindsight; the publication that mattered at the time is buried under everything published since.
Measured on a control claim: unfenced retrieval missed the pivotal publication entirely and scored 35% evidence coverage. Fenced to the claim's own year, the same query set retrieved it and coverage rose to 79%. So the fence is not only about honesty in retrospect. It is a retrieval precision feature.
Set as_of to the year a claim was made for anything that is not brand new.
What did the published evidence actually support about aducanumab and cognitive
decline at the end of 2019, using only sources available by then?Then ask the same question without the year and compare. The difference is the point.
The evidence tools send nothing to a model provider. evaluate_claim processes claim text through
a model sub-processor, named along with retention terms and hosting region in the
privacy policy. Claim text is not logged; only metadata
(the routed class, depth, counts) is kept.
This repository also publishes a thin stdio MCP client to PyPI, which predates the hosted server
and exposes an older tool set (evaluate_claim, check_evaluation, verify_attestation,
account_status). It holds no keys and runs no model; every call is proxied to the hosted engine,
and it needs a token.
Prefer the hosted endpoint above. It needs no token and carries the current tools. The client remains for existing stdio setups:
claude mcp add zetesis --env ZETESIS_TOKEN=zk_... -- uvx --from zetesis zetesis-clientTokens: https://api.zetesis.science/request-access
ZETESIS_TOKEN sets the token (verification works without one)
ZETESIS_API overrides the API base, default https://api.zetesis.science
Report vulnerabilities privately to avidan.r@zetesis.science. See SECURITY.md.
MIT licensed. The hosted engine is a separate service.
mcp-name: io.github.reutavidan/zetesis
Show the caller's Zetesis plan and remaining free evaluations or dossier credits.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, openWorldHint, and destructiveHint. Description adds specific return context (plan and credits), providing value beyond annotations without contradiction.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Single, front-loaded sentence with no wasted words; every part adds value.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a parameterless tool with an output schema, the description fully explains purpose and return values, leaving no gaps.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
No parameters; baseline score of 4 applies. Description adds meaning beyond the schema by stating what the tool shows.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description uses specific action 'show' and identifies the resource as caller's plan and credits, clearly distinguishing from sibling tools that evaluate claims or verify attestations.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No explicit when-to-use or when-not-to-use guidance. Usage is implied for checking account status, but no alternatives or exclusions provided.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get the result of a full (deep) Zetesis dossier started earlier. Returns the reading and the rendered dossier link once it is ready.
Args: job_id: the id returned by evaluate_claim for a deep evaluation.
| Name | Required | Description | Default |
|---|---|---|---|
| job_id | Yes |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already indicate readOnlyHint=true and openWorldHint=true. The description adds that the result is returned once ready, hinting at polling behavior, which is beyond annotations. No contradictions.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Description is extremely concise: two sentences plus an Args line. It front-loads the main purpose and provides necessary detail without waste.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a polling tool with one parameter and an output schema, the description covers what to expect (reading and dossier link) and how to get the job_id. Combined with annotations and output schema, it is complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema has 0% description coverage. The description compensates by explaining that job_id is the id returned by evaluate_claim for a deep evaluation, adding crucial context for correct usage.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool gets the result of a full dossier started earlier, and mentions it returns reading and rendered dossier link. It explicitly references sibling 'evaluate_claim' for job_id, distinguishing the tool's role.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description specifies that job_id comes from evaluate_claim for a deep evaluation, implying a chained usage. It does not explicitly list when not to use, but the context is clear.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Evaluate how rigorously a scientific, biomedical, clinical, or AI/ML claim was derived, and how well the evidence supports it.
Use this whenever someone wants to assess, screen, sanity-check, or do due diligence on a research claim, a study, a paper, an abstract, a preprint, a grant or a pitch: whether the methodology is sound, whether the data was harmonized and controlled properly, whether the statistics hold, whether the result reproduces, and how well the public literature, clinical trials, and filings back it. 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 screen is free and needs no token. A full attested dossier (deep) is the signed, independently re-checkable audit and needs access.
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). depth: "screen" for a fast, free pass (no token needed), "deep" for the full attested dossier (needs access). mode: "research" (science only) or "diligence" (adds the capital reading); deep only.
| Name | Required | Description | Default |
|---|---|---|---|
| mode | No | research | |
| claim | Yes | ||
| depth | No | screen | |
| context | No | ||
| documents | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already indicate readOnlyHint=true and destructiveHint=false, so the tool is safe to invoke. The description adds critical behavioral context: 'A screen is free and needs no token. A full attested dossier (deep) is the signed, independently re-checkable audit and needs access.' It also explains output includes 'sources cited by hard id (PMID, DOI, NCT, NIH grant, SEC filing)'. This goes beyond annotations and provides actionable guidance.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is front-loaded with the core purpose in the first sentence. It then provides context, use cases, and parameter details. While verbose, every sentence adds value—no filler. The structure is logical, moving from general purpose to specific parameters and modes.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the complexity (5 parameters, 1 required, 0% schema coverage) and the presence of an output schema, the description provides sufficient context: it explains all parameters, usage scenarios, and output characteristics (sources with IDs). It does not redundantly describe return values since an output schema exists. Minor gap: does not elaborate on the 'overall reading' grade format, but this is covered by the output schema.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, but the description explains each parameter inline: claim ('one or two sentences'), depth ('screen' for fast free, 'deep' for full dossier), mode ('research' or 'diligence'), context (optional background), documents (optional source text). This adds meaningful semantics beyond the schema's titles and defaults, though it could be more structured.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool evaluates how rigorously a scientific, biomedical, clinical, or AI/ML claim was derived and how well evidence supports it. It provides a specific verb ('evaluate') and resource ('claim'), with a detailed scope that distinguishes it from general search tools. Sibling tools (check_evaluation, verify_attestation) are different functions, but the description's specificity makes its purpose unambiguous.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explicitly says 'Use this whenever someone wants to assess, screen, sanity-check, or do due diligence on a research claim...' and lists concrete contexts (study, paper, grant, pitch). It distinguishes between 'screen' and 'deep' modes with cost and access implications. However, it does not explicitly state when NOT to use this tool or reference sibling tools as alternatives.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
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. No account needed.
Args: attestation_json: the full attestation object as JSON text.
| Name | Required | Description | Default |
|---|---|---|---|
| attestation_json | Yes |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and destructiveHint=false. Description adds context about what is verified (claim, evidence, conclusion) and that no account is needed. No contradictions but does not disclose additional behaviors beyond annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences plus an Args line. All front-loaded: purpose first, then usage, then parameter. Every sentence adds value. No redundancy.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has 1 simple parameter, annotations provide safety profile, and description covers purpose, usage, and parameter. Output schema exists (not shown). No gaps for the given complexity.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 0%, so description carries full burden. Description explains 'attestation_json' as 'the full attestation object as JSON text,' adding meaning beyond the schema's type and title. Single parameter well-documented.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb 'verify' and the resource 'attestation', with specific details about confirming integrity of claim, evidence, and conclusion. It distinguishes from siblings like evaluate_claim and check_evaluation.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Explicitly states when to use: someone has a Zetesis dossier or attestation and wants independent re-check. Also notes 'No account needed,' implying no authentication required. Could be improved with explicit alternatives or when-not-to-use.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.1account_statuscheck_evaluationevaluate_claimverify_attestationScored across 4 tools
Each tool serves a distinct function: creating evaluations, retrieving deep dossiers, verifying attestations, and checking account status. No overlap in purpose.
Most tools follow a verb_noun pattern (evaluate_claim, check_evaluation, verify_attestation), but account_status is a noun_noun phrase, creating a minor inconsistency.
4 tools is appropriate for a focused service like claim evaluation. It covers the essential operations without being too sparse or excessive.
Core lifecycle of evaluation creation, retrieval, and verification is covered, but missing a tool to list or manage past evaluations, which may be needed for workflow continuity.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Evidence-backed crypto due diligence with sources, freshness, and a runtime receipt on every call.
Verify stock theses against public filings with a source-backed evidence ledger.
Verify claims with verdict, confidence & cited sources; batch verify, source checks, daily brief.
Verify AI outputs against real sources. Returns a trust signal and permanent doc number.
