Discover the Best AI Tools

A hosted ingredient and product checker with condition-based flags, curated interactions, claims comparison and citations when available. Anonymous access is limited; higher metered limits use a key. Missing flags do not establish safety, and uncovered claim text may enter a private review queue, so personal or medical records should not be submitted.

A PodQuery-derived MCP tool that archives Glooko data locally and exposes summaries, charts and device-dependent analysis based on observed fields. Real end-to-end testing is documented for one pump/CGM setup; bundled sample data is synthetic. Sharing outputs with an assistant is separate from local storage, and analyses must not independently change therapy.

Unifies terminology search, code details and mapping aids across ICD-11, LOINC, RxNorm, MeSH and other sources with provenance. Source licenses apply separately from the MIT code license; SNOMED requires licensed data and self-hosted Snowstorm. Some mappings use text search and must not be treated as authoritative code conversions.

A read-only Oura API v2 connector returning compact sleep, readiness, activity and other wearable summaries for an existing assistant. Requires an Oura subscription and the user’s OAuth app for live data; demo mode is synthetic. It has no model of its own, and correlations in wearable data do not establish medical causes.

Read-only tools for Taiwan NHI billing codes, payment rules, audit clauses and related public references via OPDSTAR. Anonymous access is rate-limited. It is independently operated, not an official NHI service; regional coverage and effective dates matter, and missing results do not mean no rule exists. Diagnosis and prescribing are outside its scope.

Exposes a configured DICOM archive to assistants for study/series search, metadata, structured or PDF reports and rendered image frames. Requires Node.js 20 or later and a server implementing the needed QIDO-RS/WADO-RS endpoints, plus its credentials where applicable. Retrieval and rendering do not perform a validated diagnosis.

Combines DICOM reference lookup and HL7 parsing with licensed mapping, FHIR conversion and optional PACS connectivity. Premium tools require a license and PACS dependencies/configuration. C-MOVE is disabled by default; patient-data access and generated mappings require organizational review, and local execution does not prevent disclosure to a connected cloud model.

Provides medical-code lookup, billing validation, reimbursement references and public-health data through a hosted API. Some basic tools are anonymously rate-limited; advanced operations require an API key and credits. Use de-identified inputs and review billing suggestions against current official sources; predictions do not guarantee claim acceptance.

Connects assistants to the Lecka sports-nutrition platform for product nutrition lookup and race-fueling calculations based on duration, intensity and athlete inputs. Recommendations are tied to the Lecka catalog; the repository does not establish individualized clinical validation.

A read-only MCP server for authorized Practice Fusion FHIR records, including patients, medications, labs and bounded summaries, with audit logging and a synthetic demo mode. Requires an Open FHIR backend-services app and credentials for live access. It cannot schedule or write records; logs contain patient identifiers, summaries omit fields, and deployment compliance remains the operator’s responsibility.

A stdio-to-HTTP bridge to Vitamin D Explorer for sun times and model-based vitamin-D windows or session estimates. Public tools need no account; saved profiles/history require a separate authenticated endpoint because this bridge does not implement OAuth. Estimates omit factors such as sunscreen and individual variation, are not measurements or medical advice, and should not guide intentional burning.

A Python SDK for connecting healthcare AI to FHIR sources, validating typed resources and terminology, and exposing agent tools through MCP or LangChain. Supports gateway scaffolding and configurable deployment controls. Live EHR access requires provider authorization; schema validation does not establish clinical correctness, and governance/compliance automation remains on the roadmap.

A TastyAPI connector that submits food descriptions or publicly accessible image URLs for estimated nutrients, allergen labels and dietary flags. Requires Node.js 18 or later, an API key and paid service access. Analysis is remote; the repository does not substantiate its advertised speed or accuracy comparisons, and returned allergen labels are not a safety guarantee.

A local-first nutrition connector for food and barcode lookup, assisted meal estimates, intake and hydration logs, goals and summaries. Logs are local, while live lookups may contact USDA or Open Food Facts; fixture mode uses bundled demo data. Photo estimates need food/portion confirmation and explicit save intent, and are not precise nutritional measurements.

Connects agents to Dexcom or LibreLink Up glucose data for windowed summaries and meal-response analysis. Without provider credentials it returns explicitly marked synthetic data. Libre history is roughly 12 hours per read, regardless of a wider requested window. It is unofficial and must not be used for insulin or medication dosing decisions.

An unofficial beta connector for authorized Google Health API v4 data, with inventory checks, daily/weekly summaries and privacy diagnostics. Requires a Google Cloud OAuth client, API access and user consent. Credentials are stored locally, but provider requests and the connected AI still cross their own data boundaries; this is not a medical device.

MCP server for accessing personal health and biometric data via the Fulcra Life API with OAuth2 consent.

A personal genome analysis toolkit that analyzes raw DNA data across 17 categories with terminal-style single-page HTML visualization.

A server for managing and processing Apple HealthKit data.

A tool designed to assist with identifying potential health symptoms using AI.

A platform for bioinformatics and medical diagnosis focused on longevity research.

HealthchecksIO is an AI-powered health monitoring tool that helps users track their health metrics and detect potential issues early.

Health-Ai is an AI-powered tool designed to assist with symptom checking.

An AI model specialized in assisting with skin cancer diagnosis and analysis.