How Klarovel builds your protocol
A four-step process that takes your health data and transforms it into an evidence-based, personalized peptide protocol.
The Health Assessment
A 10-minute questionnaire that captures what a specialised longevity clinician would ask in a first consult. Every question has a specific purpose in the engine downstream, and every answer routes to a concrete rule: a contraindication check, a dose modifier, a peptide unlock or block, a monitoring trigger. There are no decorative questions.
Health conditions
Screens contraindications and ensures no peptide recommendation conflicts with an existing diagnosis. The engine carries a documented matrix that hard-blocks specific peptides for cancer history (pro-angiogenic risk), pregnancy or active conception attempts, severe cardiovascular disease, and uncontrolled diabetes. Blocks are surfaced with the reason in your protocol, not silently dropped.
Medications
Runs drug-interaction checks against your current prescriptions. The engine adjusts GLP-1 doses when stacked with insulin or sulfonylureas, flags MAO-inhibitor combinations with serotonergic peptides like Selank or Semax, and de-prioritises pro-angiogenic peptides when the user is on a chemotherapy regimen. The interaction list is sourced from FDA labels and the published peptide research catalogue.
Symptoms
Targets specific peptides against your reported complaints. Sleep architecture problems route to DSIP and the GH-axis pulse pattern; persistent inflammation routes to BPC-157 and TB-500; cognitive fog with a stress component routes to Selank and Semax. Symptom scores feed into per-peptide rankings, not just category selection.
Lifestyle
Calibrates the dose envelope based on activity level, sleep duration and quality, stress load, and nutrition pattern. A user running five training sessions per week with poor recovery is a different dose envelope from a sedentary user on the same body weight. The lifestyle layer also drives the timing recommendations, for example pre-bed dosing for the GH-axis stack when sleep is the bottleneck.
Goals
Selects the right peptide categories: recovery, cognitive, longevity, body composition, immune, or sexual health. Each goal carries its own evidence-weighted ranking against the catalogue, and the engine balances multi-goal users (the common case) rather than collapsing them to a single primary goal. You see exactly which peptide was selected for which goal in the rendered protocol.
Experience level
Sets the slot cap and the dose envelope. Tier 1 (never used peptides) caps at 2 slots and starts at the conservative end of the published range; tier 2 (beginner or intermediate) caps at 3 slots and uses standard doses; tier 3 (experienced) caps at 4 slots and unlocks advanced patterns. A GHRH plus GHRP pair counts as one slot, not two.
The Rules Engine
A deterministic 8-step pipeline that evaluates your health profile against the curated peptide catalogue. No randomness, no chatbot prompting, no probabilistic guessing. The same inputs produce the same protocol every time, and every decision is logged so the output is reproducible and reviewable.
Contraindication screening
Health conditions and medications are cross-referenced against every peptide in the catalogue. Hard blocks for cancer history apply to all GH secretagogues and TB-500; MK-677 is hard-blocked at HbA1c above the pre-diabetic threshold; GH-axis peptides are blocked during pregnancy or breastfeeding. Each block is surfaced in your protocol with the specific reason, so you understand exactly which peptide was excluded and why.
Biomarker analysis
If you upload bloodwork, the engine parses every value via OCR, normalises units across the major Norwegian and international labs, and compares against age- and sex-adjusted reference ranges. IGF-1 is converted to a sex- and age-normalised SDS score that drives the GH-axis dose envelope; HbA1c, fasting glucose, and fasting insulin gate MK-677 and adjust GLP-1 monitoring; lipids and liver enzymes flag stacking caution. Bloodwork is optional; without it, the engine uses conservative population defaults rather than fabricating numbers.
Peptide selection
Every catalogue peptide is scored against your goal profile on seven dimensions: goal relevance, safety clearance, biomarker alignment, experience-tier match, delivery-preference match, evidence-tier weight, and (for cycle 2+) continuation response from your prior cycle. The top-scored candidates fill the experience-tier slots; anything above the slot cap goes to overflow and is held for a future cycle.
Dose calculation
Doses are calibrated to your body weight, biomarker reads, experience tier, and any documented impairment (kidney or hepatic). The engine clamps doses against the published research ceiling for each peptide, so the AI layer cannot output a dose exceeding what the literature supports. GLP-1 dosing follows the FDA-label titration ladder rather than a body-weight formula.
Stacking evaluation
Peptide combinations are checked for documented synergies and conflicts. A GHRH and a GHRP collapse to a single stacking slot at full dose on both, because that is the canonical pulsatile-stimulation pattern in the research. Two GHRPs in the same protocol are removed (receptor competition). BPC-157 stacked with TB-500 is allowed but flagged with the evidence caveat.
Cycling schedules
On-cycle duration, off-cycle windows, and micro-cycling patterns are produced from the engine, not from a generic template. The output respects each peptide's published cycling research: GH-axis peptides cycle 8 weeks on / 4 off; GLP-1 weight-loss agents are continuous-use with a label-driven titration ladder; BPC-157 cycles within a 4 to 8 week window. Cycling decisions are reproduced verbatim in your daily plan.
Monitoring plans
Blood-test schedules and subjective check-in points are generated against the specific peptides in your protocol. GH-axis users get an IGF-1 recheck at week 5 and week 10; GLP-1 users get a fasting glucose plus HbA1c at week 8; users with prior hepatic or renal markers get the appropriate enzyme rechecks. Subjective tracking covers sleep quality, energy, and reported side effects so the cycle 2 follow-up has signal to work with.
Missing biomarker recommendations
If a key lab is missing and the protocol could be sharper with it, the engine tells you exactly which test to order, why it would change the recommendation, and what the rough threshold is. The protocol still ships without the missing labs, but the recommendation block surfaces what you could add to make cycle 2 more precise.
The engine makes every protocol decision. The AI layer formats those decisions into a readable report and writes the personalised reasoning sections, but it cannot change a dose, override a contraindication, add a peptide the engine did not select, or drop one it did. Every recommendation in your report is traceable back to a deterministic rule. No probabilistic guesses, no chatbot improvisation.
The Protocol Report
A comprehensive document that tells you exactly what to take, when to take it, how to prepare it, and what to monitor. Nothing is left to guesswork. The report is built in 14 sections and structured so a clinician can review it before you start a cycle.
Peptides at a glance
A summary table of every peptide in your protocol with form (vial or prefilled pen), concentration after reconstitution, dose per injection, and the syringe units to draw. Doses are written in a single unambiguous format (mg, mcg, or IU) so there is no risk of mis-drawing. The same dose appears verbatim in the daily protocol and the titration plan.
Personalised reasoning
Why each peptide was selected for you specifically, with at least two references back to your submitted data points (your goals, your sleep quality score, your training frequency, your biomarker reads, your reported symptoms). Generic copy is the failure mode this section is built to avoid, and the AI layer is forced to ground every rationale in your own questionnaire data.
Daily injection schedule
Exact timing, injection sites, and sequencing for each day of the week. Morning, pre-workout, and evening slots are populated based on each peptide's administration timing (fasted GHRP pre-bed, GLP-1 weekly on a fixed day, BPC-157 split twice daily). The schedule respects fasting requirements and prevents conflicting injection-site overlaps.
Titration plan
A three-phase structure (ramp-up, full-dose, taper) with the engine's actual titration string used verbatim, not paraphrased. For GLP-1 agents the titration is label-driven (Wegovy's 0.25 mg ladder, Mounjaro's 2.5 mg ladder); for GH-axis peptides it is a graduated start to assess tolerance; for peptides that do not require titration (BPC-157, GHK-Cu topical) the plan says so explicitly.
Reconstitution guide
Step-by-step mixing instructions for every vial peptide, with the bacteriostatic water volume calibrated to your dose for clean syringe-unit math. The guide covers needle technique, the swirl-not-shake rule, the wait-for-clear check, and the storage step. If your knowledge gaps indicate this is your first time, the guide expands into a detailed walkthrough; if you are experienced, it compresses into a reference table.
Storage instructions
Temperature requirements (lyophilised storage versus reconstituted storage), shelf life (typically 28 days reconstituted at 4 degrees Celsius for most peptides), and handling rules. Travel-day exceptions, freezer storage for unopened vials, and the visual signs that mean a peptide should be discarded are all covered.
Blood test schedule
When to test, which markers to request, and what to look for in the results. Baseline, mid-cycle (week 5), late-cycle (week 10), and post-protocol (week 16) checkpoints are populated based on the peptides you are running. Each entry includes the threshold values that would change the cycle 2 recommendation.
Post-protocol planning
Off-cycle guidance covering how long the break should last, what to expect during it, and which subjective signals are normal versus which warrant a clinical consult. The protocol also includes the placeholder for cycle 2 candidates the engine held in overflow, so you can see what is coming next before you even open the follow-up assessment.
Cycle 2 and beyond
Your protocol gets sharper with every cycle. The first cycle is calibrated against your baseline questionnaire; the second cycle is calibrated against your baseline plus everything that happened on cycle 1, which is a much richer dataset. After each round, a follow-up assessment captures what happened, and the engine adapts.
Follow-up assessment captures
- -Compliance: what you actually took versus the plan, including missed doses and the reasons.
- -Side effects: severity, duration, and how you managed them, scored on a structured rubric the engine can read.
- -Subjective improvements: sleep quality, energy, recovery, cognition, mood, body composition, broken out per goal.
- -Goal evolution: which goals are met, which still matter, which new ones have emerged.
- -Off-cycle experience: how you felt without peptides, whether benefits held or regressed.
- -Updated bloodwork: optional but recommended, especially for IGF-1, HbA1c, and any peptide-specific markers the cycle 1 plan flagged.
The engine adapts
- -Dose escalation for well-tolerated peptides that delivered (the continuation bonus is response-gated, not flat).
- -New peptide additions based on evolved goals, with the engine re-scoring the full catalogue against the updated profile.
- -Removal of peptides that did not deliver or were poorly tolerated, with the reason carried forward to the protocol.
- -Refined cycling based on your response patterns: faster ramp, longer on-cycle, shorter break, or the reverse.
- -Updated monitoring tailored to the new peptides and the trajectory your bloodwork is showing.
- -Promoted overflow candidates from cycle 1, if the slot cap kept them out the first time.
The result of this loop is that cycle 4 is dramatically more personalised than cycle 1. The deterministic engine plus a year of your own response data is a different protocol layer from any first-visit recommendation system, including the one a specialised clinician would produce on a single consult.
What the protocol is not
A Klarovel protocol is a structured document built from your submitted data, the published peptide research, and the engine's safety rules. It is explicit about what it does not do, because the failure mode in this space is overreach.
Not a prescription
Klarovel does not prescribe medication. The protocol is not a prescription, the rules engine is not a clinician, and Klarovel is not registered as a pharmacy or a medical practice in any jurisdiction. Where you live may regulate the importation, possession, or use of specific peptides; you are responsible for understanding those rules.
Not a substitute for a clinical consult
Especially for users with cardiovascular disease, active or historical cancer, type 1 or insulin-dependent type 2 diabetes, active psychiatric care, or who are pregnant, nursing, or trying to conceive. The protocol is structured precisely so it can be reviewed by a qualified clinician before you start, and we recommend that step.
Not a supply chain
Klarovel does not sell, hold, ship, or fulfil peptides. The protocol references partner suppliers who do; the order goes through them, the payment goes through them, and the regulatory posture of the fulfilment is theirs. Klarovel sits beside the transaction as the protocol layer, not inside it.
Not magic
Peptides are not stimulants, weight-loss miracles, anti-ageing cures, or general performance boosters. The catalogue is curated against research consensus, the engine respects published safety rules, and the protocol surfaces both the evidence and the gaps in the evidence. Where the literature is thin, we say so.