How larrynovsky/larry-health-os works
Larry Health OS is a proactive personal health system: it gathers medical records and personal health streams, keeps them in a local database, analyzes changes and deadlines, then reaches the person through Telegram and reminders. The principal loop runs from user-submitted data and connected sources through persistence and scheduled analysis to briefs, alerts, hypotheses, and tasks; replies and new records feed back into the system. The README documents major capabilities beyond the sampled code; edges based on those capabilities cite the README, and unsampled implementation details are not inferred.
Main entry points: scheduled.py Scheduled jobs · health_db.py Local health database · genome_intake.py Genome intake · task_agent.py Tasks and follow-up
Components 28
Data intake
- Bot intake
handlers/messages.py - Genome intake
genome_intake.py - Lab review
dashboard_routers/api_lab_review.py - Local dashboard
dashboard_routers/views.py
Health data
- Local health database
health_db.py - Lab records
labs_db.py - Genome processing
genome_pipeline.py - Health imports
import_all.py
Analysis
- Scheduled jobs
jobs/scheduled.py - Morning brief
brief_pipeline.py - Health intelligence
hai_core.py - Correlation checks
correlation_gate.py - Hypotheses
hai_hypotheses.py - Tasks and follow-up
task_agent.py
Proactive delivery
- Bot actions
bot/actions.py - Notifications
notify.py
Internal platform
- LLM client
llm_client.py - Time seam
_time_inject.py - System configuration
config_db.py - Interface language
i18n.py
Other components
- Person
- Telegram
- Health sources
- Medical documents
- Wearables
- Genome files
- macOS Reminders
- LLM service
Connections 33
- Person → Telegram: sends data
- Telegram → Bot intake: routes replies
handlers/messages.py - Bot intake → Local health database: records health data
handlers/messages.py - Person → Medical documents: provides records
- Medical documents → Local health database: ingests records
- Health sources → Wearables: supplies measurements
- Wearables → Health imports: imports data
- Health imports → Local health database: stores data
- Person → Genome files: provides file
- Genome files → Genome intake: uploads file
- Genome intake → Genome processing: dispatches processing
genome_intake.py - Genome intake → Local health database: records data
genome_intake.py - Person → Local dashboard: reviews data
- Local dashboard → Lab review: reviews labs
dashboard_routers/api_lab_review.py - Lab review → Lab records: reads and updates
dashboard_routers/api_lab_review.py - Local health database → Scheduled jobs: feeds analysis
jobs/scheduled.py - Scheduled jobs → Morning brief: schedules brief
- Morning brief → Health intelligence: requests analysis
brief_pipeline.py - Health intelligence → LLM client: uses client
llm_client.py - LLM client → LLM service: sends requests
llm_client.py - Scheduled jobs → Correlation checks: checks correlations
- Scheduled jobs → Hypotheses: runs consilium
- Scheduled jobs → Tasks and follow-up: creates follow-up
jobs/scheduled.py - Tasks and follow-up → Local health database: stores tasks
task_agent.py - Scheduled jobs → Telegram: delivers brief
- Scheduled jobs → Notifications: sends alerts
jobs/scheduled.py - Notifications → Telegram: delivers messages
notify.py - Person → Bot actions: answers prompts
- Bot actions → Bot intake: routes replies
handlers/messages.py - Tasks and follow-up → macOS Reminders: syncs reminders
task_agent.py - Scheduled jobs → Time seam: reads current time
jobs/scheduled.py - Scheduled jobs → System configuration: reads settings
jobs/scheduled.py - Bot actions → Interface language: localizes replies
bot/actions.py