Understand your health.
A useful health system should not only tell you what to do. It should teach you why, one idea at a time, and show what each recommendation rests on. Care fixes what is broken. Learning is how you improve what isn’t.
Twelve lessons. A way to reason, not a protocol.
Each lesson stands alone and ends with a check. Together they teach how to read your own evidence, weigh a recommendation and run a change without fooling yourself. Foundations release v1.0.0, frozen and versioned, so what you learned stays what you learned.
Define health in your own terms
Now, next and later: what you want to stay able to do, and what evidence would help.
Separate evidence, guidance, interpretation and action
“The report flags this” is not “you have a disease”. Keep the four apart.
Understand a measurement before comparing it
Reference range, diagnostic threshold, decision threshold, personal target: four numbers that differ without contradiction.
Build a cardiometabolic risk picture
Blood pressure, ApoB, Lp(a), glucose over months, kidney, sleep, family history. Not one score.
Reason about prevention and screening
Who a test is for, when, how often, and what a positive result actually means in your population.
Include brain, sleep and emotional health
A sleep score is weaker evidence than a timeline of opportunity, duration, symptoms and daytime function.
Treat physical capacity as several trainable qualities
Endurance, strength, power, balance, mobility. A step count measures exposure, not fitness.
Evaluate nutrition by pattern, adequacy, response and sustainability
An explicit goal over an ideology, judged by outcomes and burden.
Reconcile medications and supplements as interventions
Listed, prescribed, dispensed, actually taking: four different facts.
Run measured actions without fooling yourself
Baseline, one change, a review date and a rule for what counts as success, set before you start.
Maintain a useful longitudinal view
A sparse honest view beats a complete-looking one filled with assumptions.
Build a qualitative prevention-priority review
The longevity review: rank the work, keep the unknowns visible, never compress it to one number.
One bounded idea at a time, then a check.
Ask “what can you teach me?” and it shows the map, asks what outcome matters to you, and starts small. Ask a direct question and it answers the question; it does not force a lesson. A detour into a deeper unit never loses your place in the course.
Presented is not completed. Completed is not mastered. HealthOS records which one actually happened, and a plan drawn up in a lesson becomes a real action only when you say so.
Forty-six optional units across twelve topics.
For the questions the foundation raises: liver fat versus fibrosis, conflicting lab results, what an aging clock demonstrates. Pick the question and it selects the smallest unit that answers it, each with its sources and a check with an expected answer.
Longevity standards, not lab ranges. With the sources showing.
The default stance is preventive: preserve long-term function, address important preventable risks early, connect every measurement to a decision. It is distilled from public-health and clinical sources and from clearly attributed longevity lenses, including Peter Attia’s and Bryan Johnson’s. Neither is affiliated with or endorses HealthOS.
Know what is evidence, interpretation and uncertainty.
HealthOS keeps source material close to the decisions it informs, so you can go deeper without losing the thread.
Keep the source
Preserve the record, date, publisher and context behind each claim or measurement.
Keep claims in lane
Separate what a source observed, what it inferred and what HealthOS interprets for you.
Show what changed
Track revisions when new evidence, goals or constraints change the plan.
Make uncertainty useful
Name gaps, conflicts and confidence limits so the next measurement has a purpose.
Covered is not the same as learned.
HealthOS is building a source-indexed corpus, held to longevity standards, across cardiovascular and metabolic health; cancer and screening; brain, sleep and mental health; strength, fitness and bone; and immune health. Its learning ledger records what was explained, what you understood and what comes next. Compatible agents can use the same persistent context; gaps and corrections remain visible.