@almostjacked/fitness-tools

New here? Start with the getting-started tutorial — zero to an understood result in one page. The reference below is the fast version.

Validated, self-describing fitness calculators for TypeScript — runs natively in the browser and on the server. BMR/TDEE, body fat, 1RM, macros, activity multiplier, powerlifting attempts, and natural muscular potential, each with multiple methods and a consensus across them.

· Types ✓ · 1 dependency (zod) · MIT

npm i @almostjacked/fitness-tools

Quick start

import { REGISTRY } from "@almostjacked/fitness-tools";

const tdee = REGISTRY.get("tdee")!;
const out = tdee.compute(tdee.input.parse({
  sex: "male", age: 30,
  height: { value: 180, unit: "cm" },
  weight: { value: 80, unit: "kg" },
  activity: "moderate",
}));
// out.results   → [{ method: "mifflin", value: 2759, ... }, { method: "harris", ... }]
// out.consensus → { mean: 2816.05, median: 2816.05, min: 2759, max: 2873.1, n: 2 }

Prefer raw functions? They're exported too, fully tree-shakeable:

import { mifflinBmr, activityMultiplier } from "@almostjacked/fitness-tools";
mifflinBmr("male", 80, 180, 30) * activityMultiplier("moderate"); // 2759

Validated & self-describing

Every tool carries a Zod schema, so a single definition gives you runtime validation, static types, and a JSON-Schema description — for one small dependency. TypeScript can't catch a negative age or a body-fat of 200% at an untrusted boundary like a form; the schema can:

const r = tdee.input.safeParse({ sex: "male", age: -5, /* ... */ });
if (!r.success) {
  // r.error.issues → structured, per-field errors you can render in a form
}

The same schemas describe each tool (tool.input / tool.output), which is how the companion HTTP server generates its catalog and OpenAPI spec — no hand-written docs.

Runs in the browser, no build step

<script type="module">
  import { mifflinBmr } from "https://esm.sh/@almostjacked/fitness-tools";
  console.log(mifflinBmr("female", 65, 168, 28));
</script>

The package is isomorphic — zero Node-only APIs — so the identical code runs in a browser, an edge function, or Node.

Tools

| Tool | Methods | Description | |---|---|---| | tdee | mifflin, harris, katch, cunningham | Estimate BMR and TDEE via Mifflin-St Jeor, Harris-Benedict, Katch-McArdle, and Cunningham. Provide body_fat or lean_mass to unlock the LBM-based methods. | | body-fat | navy, jackson-pollock-3, deurenberg | Estimate body-fat % via US Navy circumference, Jackson-Pollock 3-site skinfold, and Deurenberg (BMI-based) methods. | | one-rep-max | epley, brzycki, lombardi, wathan, oconner, mayhew | Estimate 1RM from a submaximal set via Epley, Brzycki, Lombardi, Wathan, O'Conner, and Mayhew; returns a %1RM load chart. | | macros | g-per-kg | Compute protein/fat/carb grams for a calorie target using the g-per-kg-bodyweight method, with goal-based protein defaults. | | activity-multiplier | lookup, neat-eat | Estimate the TDEE activity multiplier via the classic lookup table or a NEAT+EAT model (occupation/steps for non-exercise, training volume for exercise). | | powerlifting-attempts | standard | Deterministic opener/second/third attempts from an estimated 1RM, plus a warmup ramp and per-side plate loading. Tunable by aggressiveness and available plates. | | muscle-potential | casey-butt, ffmi-cap, berkhan | Estimate drug-free maximum bodyweight at a target body-fat % via Casey Butt (wrist+ankle), the FFMI~25 natural cap, and Berkhan's max-contest-weight model. Men only in v1. | | ffmi | standard | Compute the Fat-Free Mass Index (FFMI) and its height-adjusted form from weight and body fat (or lean mass), and flag whether it exceeds the ~25 natural ceiling. | | rsmi | direct, wen-2011 | Estimate the appendicular skeletal muscle index (RSMI) from a DXA value (direct) and/or an anthropometric estimate (Wen 2011), and flag low muscle mass against EWGSOP2, AWGS, and Baumgartner sarcopenia cutoffs. |

Each tool runs several methods and reports a consensus (mean/median/min/max/n) across them; methods whose inputs are missing are listed in skipped (or raise in explicit mode). Units are explicit everywhere ({ value, unit }).

Need it over HTTP?

There's a reference HTTP server — @almostjacked/fitness-tools-api — that re-exposes these calculators over HTTP for clients that can't npm install a TS package (other languages, curl, no-code tools). It's optional: if you're in JS/TS, use this package directly. The server adds a network boundary, not capability.

License

MIT — free, open-source, and usable in commercial products. See LICENSE. Built on published formulas (Mifflin-St Jeor, US Navy, Jackson-Pollock, Epley, Casey Butt, etc.); contributions welcome.