"How to implement intent recognition in MCP" matching MCP servers:

Enables AI agents to rank US healthcare organizations for clinical trial site feasibility, predicting patient accrual for a Graves' disease protocol using claims and registry data.

Enables natural-language querying of a mock legacy healthcare database and returns validated FHIR resources (Patient, Observation, Condition).

Enables AI assistants to read COVID-19 lateral flow test results from photographs by calling a deployed computer-vision pipeline, returning verdicts like positive or negative with confidence.

MyChart MCP Server

An MCP server that grounds protein research in the UniProt SPARQL endpoint, providing tools for querying proteins, sequences, variants, diseases, and more via intent-named tools and raw SPARQL.

Explains why two scientific papers disagree. A contradiction explorer for bench scientists that compares two studies' designs and grounds every claim in a verbatim quote from the source text.

Grounds gene-nomenclature work in the HUGO Gene Nomenclature Committee (HGNC) dataset, enabling resolution of gene symbols and IDs to canonical HGNC identifiers, plus cross-references and batch operations.

An MCP server implementing the ROBINS-I V2 framework for risk-of-bias assessment in non-randomized studies, with deterministic algorithms and full provenance. It enables users to parse study documents, specify target trial results, answer signalling questions with evidence-bound quotes, and compute or override domain judgements.

TransBench is a translational research agent that takes free-text clinical observations, decomposes them into biological axes, generates falsifiable hypotheses grounded in PubMed evidence, and designs reproducible computational experiments with concrete datasets for Claude Science.

Monitors PubMed and medRxiv for configurable research topics, summarizes papers using a local LLM, and stores deduplicated results in a local database.

Enables Claude Desktop to query and analyze diabetes data from Omnipod 5 and Glooko through natural language, providing clinical audit and triage insights such as time in range, trends, and bolus analysis.

Enables validation of clinical notes in SOAP format, checking required fields and semantic coherence using a local LLM, all processed locally with no data leaving the machine.

Renders interactive 2D molecular structure diagrams from SMILES notation and computes molecular properties like molecular weight, LogP, and TPSA, directly in the chat.

Search 200M+ patents and scientific literature in natural language using PatSnap's global R&D intelligence database.

Enables AI assistants to answer healthcare REIT portfolio questions with deterministic, provable answers grounded in Omega Healthcare 10-K filings and CMS nursing home data, including status, caveats, and provenance.

Enables a dental office's AI agent to verify benefits for a batch of up to 25 patients in a single MCP call, returning coverage percentages, frequency limits, waiting periods and deductible/annual-maximum remaining, each tied to a plan-clause citation. Exceptions such as identity mismatches or ambiguous cases are routed to a human review queue rather than guessed, with every call audit-logged.

Provides phenotype-aware cardiovascular triage tools including FHIR R5 case retrieval, missing-data analysis, ECGFounder-backed ECG interpretation, risk summarization, and structured SBAR handoff in multiple languages.

Enables LLM agents to search biomedical literature, retrieve article details, find related papers, manage caching, and download open-access full texts through MCP with structured, agent-friendly responses.

Enables checking whether a medicine has a reported shortage in Portugal (Infarmed monitoring, export-suspension and prior-notification lists) and finding pharmacies in a municipality that are on duty today or open now, with addresses, phone numbers and distances. Also surfaces the relevant help lines and next steps, read-only from public sources and without pharmacy stock data.

Provides safe read-only access to synthetic healthcare data via FHIR R4, enabling AI assistants to search patients, retrieve vitals, and summarize conditions using a public sandbox.

Enables MCP clients to annotate and classify single human genetic variants in any notation, returning an auditable points-based ACMG verdict with its criteria, gnomAD frequencies, in-silico predictors, ClinVar data, and per-transcript consequences. It is read-only, requires no account or API key, and keeps answers compact while linking back to the source.

Enables AI agents to search Iranian doctors, hospitals and clinics by specialty, city, name, gender and online visit, then read profiles, ratings, insurances, patient reviews, visit prices in Toman and free appointment times. Read-only, with no login, booking or payment, so users compare options and get a profile link to book themselves.

SuperGlookoQuery is a Claude Desktop extension (MCPB) that connects to Glooko diabetes device data and exposes it as clinical analysis tools (time in range, GMI, trends, enriched bolus/basal analysis, charting), for any pump/CGM combination Glooko supports rather than one fixed device. It's a fork of Richard Hall's podquery-mcp, reworked to discover each account's own data shape rather than assumi

A Model Context Protocol (MCP) server for searching the National Provider Identifier (NPI) registry. This server provides tools to search and retrieve information about healthcare providers and organizations in the United States.

Enables conversational access to 400,000+ clinical trials on ClinicalTrials.gov, allowing users to search, compare, and retrieve trial details using plain English through MCP-compatible clients like Cursor and Claude Desktop.

Enables AI assistants to upload clinical documents and medical device images for health data extraction, normalization, and retrieval of structured coded results in formats such as JSON, FHIR, HL7, CSV, and Parquet.

Enables AI agents to read and analyze Dexcom CGM data through tools for current glucose, recent historical readings, and glucose statistics such as time in range.

Translates HL7 v2.x messages to FHIR R4, validates against profiles like US Core, and explains errors in natural language for AI agents.

Read-only FHIR chart tools for agents, with opt-in human-approved write proposals through MCP elicitation.

A TypeScript MCP server for interacting with CDISC Define-XML 2.0 and 2.1 files, enabling users to load, search, and retrieve clinical data metadata including datasets, variables, and codelists. It supports local file processing with document caching and provides responses in both JSON and Markdown formats.