# thetahealth/mirobody

The AI-native health data engine — collect, standardize, and reason over labs, wearables & genomics.

Repository: https://github.com/thetahealth/mirobody
Canonical: https://ross.abutalabs.com/products/mirobody
Homepage: https://mirobody.ai
Language: Python
License: Apache-2.0
License Family: permissive
Topics: fhir, mcp, ai-agents, genomics, health-data, lab-reports, loinc, python, self-hosted, terminology, wearables
Last push: 2026-09-02T08:44:33+00:00

## Health v2 (maintenance only)
Score: 84/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 100, release rhythm 100, longevity 19
- inputs: {"age_days": 277, "days_push": 0, "days_rel": 2, "gap_med": 3.5, "n_releases_24m": 3}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1229, forks 216 (observed 2026-09-03T02:15:18.595230+00:00)

## What it is
Mirobody is an open-source, AI-native health data engine that collects readings from lab reports, wearables, and genomics, standardizes them into canonical codes (LOINC, SNOMED CT, RxNorm), and lets AI agents reason over the results. It ships as a Python package with a CLI, an OpenAI-compatible API, and MCP support, and can be self-hosted or used via Mirobody Cloud.

## Use cases
- parse lab report PDFs into standardized health indicators
- resolve messy health measurement names to LOINC or SNOMED codes
- unify wearable and Apple Health data into one record
- build an AI agent that answers questions about my blood tests
- normalize health data units and reference ranges across sources
- self-host a health data pipeline for genomics and labs
- expose personal health records to LLMs via MCP

## When to choose
- you need to standardize fragmented health data (labs, wearables, genomics) into canonical clinical codes
- you want AI agents to reason over original health documents with citations
- you need FHIR-aligned, self-hosted health data infrastructure
- you want an OpenAI-compatible API for health data queries

## When to avoid
- you need a turnkey consumer health app with UI rather than a data engine
- you require HIPAA-compliant hosting out of the box without operating it yourself
- your project has nothing to do with health or biomedical data

## Facets
- artifact type: library
- maturity: active
- function: etl, nlp, agent-framework, mcp, data-science, sdk
- domain: healthcare, artificial-intelligence
- platform: python, self-hosted, cli
- tags: health-data, fhir, loinc, snomed, wearables, lab-reports, genomics, medical-terminology, indicator-resolution, ai-agents, data-engineering, natural-language-processing

## Member repositories
- thetahealth/mirobody (main) score 84

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:18.595230+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:57:04.309009+00:00, confidence not recorded.
  - readme: https://github.com/thetahealth/mirobody (fetched 2026-09-03T02:15:18.595230+00:00, sha 0b575d7f40c5)
  - homepage: https://mirobody.ai (fetched 2026-08-29T12:59:14.678346+00:00, sha 88c238ddd4be)
  - site_page: https://docs.mirobody.ai/ (fetched 2026-08-29T12:59:14.687709+00:00, sha 3a1449d6ad47)
  - registry_pypi: https://pypi.org/pypi/mirobody/json (fetched 2026-08-29T12:59:14.689699+00:00, sha 066fbe08148c)
- Data as of 2026-08-30T08:39:29.467469+00:00.