Free medical evidence tools: graded CliniAtlas answers, PubMed metadata and EU SmPC links.
Scored across 7 tools
Tools are mostly distinct: search_evidence_library, search_medical_evidence and ask_cliniatlas are differentiated by published-vs-live-vs-generated workflow, and get_evidence_answer vs get_cliniatlas_answer are separated by slug vs job_id. Minor potential confusion remains between the two search tools and the two get tools, but descriptions clarify the intended sequence.
All names use snake_case with a consistent verb-first pattern (ask_, get_, lookup_, search_). Prefixes are predictable and resource names are clear, with no mixed casing or vague verbs.
Seven tools map cleanly to the evidence workflow: library search, published answer retrieval, live search, generated answer, async result, PubMed lookup and EMA lookup. No obvious bloat or missing coverage due to count.
The surface covers the main evidence lifecycle: search published/live literature, generate and retrieve cited answers, and look up PubMed/EMA specifics. Minor gaps exist—e.g. no direct abstract or full-text retrieval and limited medicine search—but agents can follow provided URLs or use described workarounds.
Full CliniAtlas answer to a clinical or research question, with numbered citations.
Searches the literature, guidelines and EMA documents and writes a cited answer, the
same answer cliniatlas.com gives. Uncited parts are reported through grounding.
Check search_evidence_library first; when a published answer matches, use that
instead, because each call here uses one of the user's free daily answers (see
free_answers_remaining_today). Usually takes 35 to 50 seconds: if status is
"running", call get_cliniatlas_answer with job_id. Do not include patient
identifiers. For literature research and education, not individual care decisions.
| Name | Required | Description | Default |
|---|---|---|---|
| locale | No | ||
| question | Yes |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Beyond the readOnly/destructive annotations, it discloses latency (35-50 seconds), the async status contract ('if status is "running", call get_cliniatlas_answer with job_id'), a cost side-effect (consumes one free daily answer), a privacy constraint (no patient identifiers), and grounding behavior for uncited parts. This is rich operational context the annotations cannot convey.
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?
Front-loaded with the purpose, then sequenced into alternatives, cost, async behavior, privacy, and scope. Each sentence carries actionable information with no filler; nothing could be removed without losing decision-relevant content.
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?
With an output schema present, return values need not be described. The description still covers the full operational picture an agent needs: when to prefer an alternative, how long to wait, how to poll, cost, privacy limits, and intended scope.
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% and the description does not compensate. The 'question' parameter is only implied by 'a clinical or research question', and 'locale' is never mentioned at all, nor are the length bounds (8-2000 chars). For a 2-param tool with no schema docs, this leaves a real gap.
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?
States a specific verb and resource ('Full CliniAtlas answer to a clinical or research question, with numbered citations') and enumerates the sources searched (literature, guidelines, EMA documents). It clearly distinguishes itself from siblings like search_evidence_library and get_cliniatlas_answer, which are named explicitly.
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 routes the agent: 'Check search_evidence_library first; when a published answer matches, use that instead,' with the reason (each call consumes a free daily answer). It also names the async continuation path (get_cliniatlas_answer) and states an exclusion: 'For literature research and education, not individual care decisions.'
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Collect the answer for a job_id returned by ask_cliniatlas.
Waits up to about 40 seconds for the result. If status is still "running", call it
again with the same job_id.
| Name | Required | Description | Default |
|---|---|---|---|
| job_id | Yes |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already establish read-only, idempotent, non-destructive, and non-open-world behavior. The description adds useful async-specific behavior: the ~40 second wait and the need to poll again if the status remains running.
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 two short sentences with no wasted text. The core instruction is front-loaded, and the polling caveat follows naturally.
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?
With an output schema present, the description need not explain return values. Annotations cover the safety profile, and the description supplies the essential async polling behavior, making this complete for its scope.
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%, so the description must add meaning for the single parameter. It does so by identifying job_id as the value returned from ask_cliniatlas, which is more valuable than repeating the schema pattern.
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 states a specific action (collect the answer) tied to a specific resource (a job_id returned by ask_cliniatlas). It clearly distinguishes this polling/retrieval tool from the initiating sibling ask_cliniatlas.
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?
It tells the agent to use a job_id returned by ask_cliniatlas and explains the polling pattern: wait roughly 40 seconds, and call again with the same job_id if the status is still running. It lacks explicit when-not guidance or comparison to other retrieval siblings, but the usage context is clear.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Full published answer: cited paragraphs, limitations and graded sources.
Use a slug from search_evidence_library. Citation markers like [1] refer to the
numbered sources. Show the sources and intended_use with the answer.
| Name | Required | Description | Default |
|---|---|---|---|
| slug | Yes | ||
| locale | No |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnly, idempotent, non-destructive, so the safety profile is covered. The description adds non-obvious context: citation markers like [1] map to numbered sources, and the agent should surface sources and intended_use alongside the answer — real behavioral guidance beyond the structured fields.
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?
Three short sentences, front-loaded with what is returned, followed by sourcing and citation handling. Every sentence carries information; the only mild inefficiency is the imperative 'Show the sources...' which addresses the agent rather than describing the tool.
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?
An output schema exists so return structure need not be restated, and the description still names the key content pieces (cited paragraphs, limitations, graded sources) plus the citation-marker convention. The only gap is locale, which is neither described nor disambiguated anywhere.
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%, so the description carries the full burden. It explains where slug comes from (search_evidence_library) but gives no format details, and the locale parameter is never mentioned, leaving one of two params completely undocumented.
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?
States a specific verb (get) and resource (full published evidence answer) and enumerates the payload — cited paragraphs, limitations, graded sources — which distinguishes it from the sibling search_evidence_library. It stops short of an explicit 'this returns one document by identifier' framing but is clearly more than vague.
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 line 'Use a slug from search_evidence_library' establishes the precondition and the correct upstream tool, giving the agent a clear workflow context. There is no explicit when-not guidance or a named alternative for other lookup paths, which keeps it below a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Centrally authorised EU medicines by brand name or active substance.
Returns the EMA summary, authorisation status and ema_url, the EMA page whose
Product information section holds the SmPC. Nationally authorised medicines are not
in this register; check the national agency for those.
| Name | Required | Description | Default |
|---|---|---|---|
| name | Yes | ||
| limit | No |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnly, idempotent, non-destructive and closed-world, so the safety profile is covered. The description adds useful behavior beyond that: it names the returned fields (EMA summary, authorisation status, ema_url) and points to where the SmPC lives on the returned page.
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 tight sentences, both earning their place, with the scope and the exclusion front-loaded. Slightly telegraphic alongside the title but not wasteful.
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?
An output schema exists, so return values need not be spelled out, yet the description still usefully flags the key fields. Combined with the annotation coverage and the name-vs-substance note, an agent has enough to call the tool correctly.
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?
With 0% schema description coverage the description must carry the param burden, and it does explain the key 'name' parameter by stating it accepts either a brand name or an active substance. The 'limit' parameter is left entirely undiscussed, so compensation is only partial.
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?
Names the specific resource (centrally authorised EU medicines) and the lookup key (brand name or active substance), and explicitly excludes nationally authorised medicines. It is clearly separable from lookup_pubmed and search_evidence_library, though it relies on the name/title for the verb 'lookup' rather than stating an action directly.
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?
Provides a concrete when-not: nationally authorised medicines are absent from this register and the agent should consult the national agency instead. This is a real selection constraint, but it does not route the agent to any of the sibling tools for adjacent query types.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Citation metadata for a PubMed ID: title, journal, year, authors, DOI and design.
Abstracts are not returned; follow pubmed_url to read them.
| Name | Required | Description | Default |
|---|---|---|---|
| pmid | Yes |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnly, idempotent, openWorld and non-destructive, so the safety profile is covered. The description still adds a genuinely non-obvious behavioral fact: abstracts are excluded and the agent should follow pubmed_url to read them. That is real value beyond the structured fields, though pagination/error behavior is unaddressed.
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 short sentences, front-loaded with the return payload and followed by the one caveat that matters. Nothing is redundant or padding.
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?
With an output schema present, return structure needn't be explained, yet the description still summarizes it usefully. Combined with the abstract caveat, an agent has enough to call this correctly; only sibling routing and parameter format are thin.
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%, so only the regex pattern documents the single pmid argument. The description partly compensates by framing pmid as a PubMed ID, but adds no format, length, or validity detail beyond that. Marginal value over the schema.
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?
States a specific verb and resource ('Citation metadata for a PubMed ID') and enumerates the returned fields (title, journal, year, authors, DOI, design), so an agent knows exactly what it gets. It does not explicitly contrast itself with siblings like lookup_eu_medicine or search_evidence_library, but the PubMed-specific scope makes the distinction reasonably clear.
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 gives no when-to-use guidance or routing versus the sibling lookup/search tools. The only constraint offered is negative scope ('Abstracts are not returned'), which is a limit, not usage guidance. An agent must infer that this is the single-record citation fetch.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Find published CliniAtlas evidence answers matching a clinical question.
Call this first for a clinical question: it is instant and does not use the user's
free daily answers. Returns question, summary and link for each match. The library
is a set of worked answers to common questions; when one matches, use
get_evidence_answer for the full answer. locale (for example de, fr, hr) returns
translated answers where they exist.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | ||
| query | Yes | ||
| locale | No |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already establish read-only, idempotent, non-destructive, closed-world behavior. The description adds real value beyond them by disclosing the quota-free, instant nature of the call and the shape of matches returned (question, summary, link). It does not, however, mention pagination or match-quality behavior.
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 short paragraphs, front-loaded with the routing decision, and every sentence carries information. The 'Returns question, summary and link' clause is mildly redundant given the output schema exists, keeping it from a 5.
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 read-only search tool with an output schema documenting return values, the description covers purpose, precedence, quota cost, and locale handling. Nothing essential to correct invocation is missing.
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%, so the description must compensate. It usefully explains locale ('de, fr, hr' returns translated answers where they exist), but leaves query and limit (including the 1-10 range and default) entirely to the schema, so compensation is only partial.
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?
States a specific verb and resource ('Find published CliniAtlas evidence answers matching a clinical question') and clarifies that the library is a set of worked answers. It does not explicitly distinguish this tool from close siblings like search_medical_evidence, so it stops short of a 5.
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?
Gives an explicit routing rule: 'Call this first for a clinical question' because it is instant and does not consume the user's free daily answers. It also names the follow-up tool (get_evidence_answer) once a match is found, so an agent knows both when to use it and what to do next.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Live search of PubMed, Europe PMC and clinical guidelines.
Returns up to eight sources for a clinical or research question, each with its study
type, evidence level and link, usually in about 20 seconds. It does not write an
answer: cite the sources by number, or use ask_cliniatlas for a full cited answer.
For an EU medicine's label or SmPC, use lookup_eu_medicine. Do not include patient
identifiers. Searches are limited per user per day.
| Name | Required | Description | Default |
|---|---|---|---|
| locale | No | ||
| question | Yes |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already cover the safety profile (readOnly, openWorld, idempotent, non-destructive), yet the description still adds genuinely new operational context: ~20 second latency, an eight-source cap, per-user daily search limits, and a PHI warning ('Do not include patient identifiers'). These are traits the annotations cannot express.
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?
Front-loaded with the core capability and result shape, then the routing rules and the safety/limits caveats. Every sentence carries distinct, non-redundant information; nothing is padding.
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 read-only search tool with a declared output schema, the description covers what an agent needs: scope, result count, latency, non-answer behavior, sibling routing, PHI prohibition and rate limits. Return-value structure is already in the output schema, so its absence from the description is not a gap.
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% for both parameters, so the schema supplies only names, types and constraints (question minLength 8/maxLength 1000, locale pattern). The description says nothing about what 'question' should contain or what 'locale' affects, leaving the semantics entirely to inference.
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?
Names the specific verb and resource (live search of PubMed, Europe PMC, clinical guidelines) and states scope: up to eight sources with study type, evidence level and link. It explicitly distinguishes itself from ask_cliniatlas (full cited answer) and lookup_eu_medicine (EU label/SmPC), so an agent can route correctly without opening a schema.
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?
Gives explicit when-not guidance: 'It does not write an answer: cite the sources by number, or use ask_cliniatlas' and 'For an EU medicine's label or SmPC, use lookup_eu_medicine.' Alternative tools and the conditions that select them are named directly.
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.
ask_cliniatlasget_cliniatlas_answersearch_medical_evidenceget_evidence_answerlookup_eu_medicinelookup_pubmedsearch_evidence_libraryPharmacogenomics for AI agents: cited drug–gene, gene and rsID lookups plus a medication screen.
Medical RAG: semantic search for clinical guidelines, drug interactions, diagnoses & EHR data.
Medical RAG: semantic search for clinical guidelines, drug interactions, diagnoses & EHR data.
Diagnoses, drugs & lab codes: ICD-11, LOINC, RxNorm, MeSH, ATC, CID-10. 33 tools, MIT.
Add one secure layer between your agents and this server.