PGxAI Andromeda

Review Medication Evidence

review_medication_evidence

Read-onlyIdempotent

Look up published drug-drug interactions (DDI), gene-drug associations (PGx) and drug-disease information (DCI). Use for general reference questions involving two or more drugs, one drug and a named disease or population term, or one drug and an explicitly supplied interpreted gene genotype or phenotype. Provide drug names in English, or a known RxNorm identifier for a drug; when the route of administration is known, give it in parentheses after the name, for example "tacrolimus (topical)" or "furosemide (oral)", using only one of: oral, topical, injectable, ophthalmic, nasal, inhalant, dental, rectal, transdermal, vaginal, otic, buccal, sublingual, irrigation, intravesical, drug implant, mucosal, pyelocalyceal, intraperitoneal; when it is not stated, send the drug name alone and do not guess a route. Send at most 20 medications per request (split a longer list into several requests); map each disease or population term with get_condition_category first and submit the most specific matching sub-code name from it (a category name only for a general topic). If genes is omitted, no gene-drug analysis is performed.

Returns a status, counts, notices and findings with source links; source_url, phenotype and implications may be null.

findings means findings were returned. checked_no_findings means the lookup completed without a returned finding; it is not proof that published evidence is absent or that use is universally safe. nothing_to_check means no medications were submitted. no_data_returned is inconclusive.

The finding severity uses a shared nominal value and is not a ranking of importance. Preserve the cited evidence and its limitations when explaining results.

Input Schema

Name	Required	Description
`drugs`	Yes	At least one medication, at most 20 per request (split a longer list into several requests). Each item requires rxnorm and/or text.
`genes`	No	Optional interpreted gene/genotype or gene/phenotype entries. Required for PGx findings; omit for DDI/DCI-only requests.
`conditions`	No	Optional disease or population names. Each entry must be an exact name from get_condition_category (never an ICD-10 code): prefer the specific sub-code when a specific disease is named, a category name only for a general topic; free text is not accepted. Age is written as the word age plus a whole number 0-120, for example "age 72" (never "72 years old" and never a literal "<N>"); sex is "male" or "female".
`confirm_no_direct_identifiers`	No	Set true ONLY when resending after this tool flagged a drugs/genes field as a possible direct identifier (ner_* pattern) and you have reviewed it and are certain it is not one (e.g. an ordinary drug/gene name the model mismatched). It is ignored unless this server flagged that exact value moments ago, so setting it up front has no effect.

Output Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedNov 2, 2026

changedInput schema / properties / drugs / description

Previous value: -"At least one medication. Each item requires rxnorm and/or text."New value: +"At least one medication, at most 20 per request (split a longer list into several requests). Each item requires rxnorm and/or text."

addedInput schema / properties / drugs / items / properties / route

Added value: +{
+  "description": "Route of administration, only when the user stated it; one of: oral, topical, injectable, ophthalmic, nasal, inhalant, dental, rectal, transdermal, vaginal, otic, buccal, sublingual, irrigation, intravesical, drug implant, mucosal, pyelocalyceal, intraperitoneal. Equivalent to writing it in parentheses after the name in text, for example tacrolimus (topical). Omit it when not stated; never guess.",
+  "type": "string"
+}

First observedOct 23, 2026

TDQS

A4/5.0

Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly, idempotent, non-destructive, closed-world, so safety needs no restatement. The description adds real value by decoding the four status values and warning that checked_no_findings is not proof of absence and that severity is nominal, but it also spends sentences on route formatting that the schema already documents, diluting the genuinely new behavioral content.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness2/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The critical scoping information is front-loaded, but the body is bloated: the entire 18-value route enumeration is reproduced verbatim even though the schema lists it twice more, and the confirm_no_direct_identifiers semantics are explained at length in prose while also appearing in the schema. Several long run-on clauses bury the two-or-more-drugs rule mid-sentence.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a read-only reference lookup with a full output schema, the description covers everything an agent needs: accepted input shapes, normalization prerequisites, batching limits, the gene-omission behavior, nullability of source_url/phenotype/implications, and a careful interpretation caveat on each status value.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so every parameter, including the nested drug object fields and the confirm_no_direct_identifiers flag, is already documented at source. The description restates the 20-drug cap, the route parenthetical convention and the confirm flag's conditional/ignored-unless-flagged behavior, but adds little semantic detail the schema does not already carry.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The opening sentence names a specific verb (look up) and three concrete resource types (DDI, PGx, DCI), which tells an agent exactly what domain question this tool answers. It is cleanly distinct from its only sibling, get_condition_category, which it names as a separate prerequisite step.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives explicit triggering shapes: two or more drugs, one drug plus a named disease/population term, or one drug plus an interpreted genotype/phenotype. It also states an exclusion ("If genes is omitted, no gene-drug analysis is performed") and routes condition normalization to get_condition_category.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All PGxAI Andromeda tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?