一键导入
meal-planner
Create personalized meal plans with macros, shopping lists, and prep guides.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Create personalized meal plans with macros, shopping lists, and prep guides.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Design static ad creatives for social media and display advertising campaigns.
Source and evaluate candidates with job analysis, CV screening, and pipeline tracking.
Find relevant companies and leads for B2B sales with ICP definition and qualification frameworks.
Draft emails, manage calendars, prepare agendas, and organize productivity.
Create brand identity kits — logos, color palettes, typography, naming, and style guides.
Perform competitive market analysis with comparisons and strategic recommendations.
| name | meal-planner |
| description | Create personalized meal plans with macros, shopping lists, and prep guides. |
Create personalized meal plans with calculated macro targets, shopping lists, and training schedules.
DISCLAIMER: General nutrition and fitness information only — not medical or dietetic advice. Users with medical conditions, eating disorder history, pregnancy, or on medications should consult a registered dietitian or physician.
User wants a weekly meal plan hitting specific macros
User needs a shopping list generated from a plan
User wants a training split paired with nutrition
User wants to adjust an existing plan (plateau, weight change, preference change)
Medical dietary needs (renal, diabetic, celiac management) → refer to RD
Single recipe creation → use recipe-creator skill
Health data analysis → use personal-health skill
When a meal in the plan is complex enough to need step-by-step instructions (e.g., "Chicken Parmesan"), the meal-planner skill defines what to eat and when; the recipe-creator skill builds the full recipe (ingredients, steps, timers, timeline). If the user's project has a recipe database, link meals to recipe IDs. If standalone, embed simplified cooking notes in the meal item.
Goal: fat loss / maintenance / muscle gain
Sex, age, height, weight
Activity level (see multiplier table below)
Dietary restrictions / allergies
These inputs have the biggest impact on adherence. Do not skip them.
| Input | Why it matters | Example |
|---|---|---|
| Eating window / fasting pattern | Determines number of meals, slot names, and calorie distribution. IF (16:8, 20:4, OMAD) users skip breakfast entirely — never plan meals outside their window. | "I skip breakfast, first meal at noon" |
| Meals-per-day preference | Some people want 3 big meals, others want 5 smaller ones. Respect this. | "4 meals" |
| Food preferences / cuisine | Build around foods the user already likes. Ask: "What are 5 meals you eat regularly and enjoy?" and "Any foods you hate?" | "I like burgers, burritos, stir-fry" |
| Cooking time per meal | Determines complexity. 15 min = sheet pan & stir-fry only. 45 min = full recipes. | "30–45 min max" |
| Budget | Affects protein source choices (chicken thigh vs. salmon) and organic vs. conventional. | "Moderate" |
| Cooking skill | Beginner = fewer techniques, clearer instructions. Advanced = more variety. | "Intermediate" |
| Household size | Affects portion scaling and shopping quantities. | "Cooking for 1" |
Do not hardcode meal names. Derive slot names and timing from the user's eating window:
| Pattern | Slots | Timing |
|---|---|---|
| Standard 3-meal | breakfast, lunch, dinner | 7am, 12pm, 6:30pm |
| Standard + snacks | breakfast, snack-am, lunch, snack-pm, dinner | 7am, 10am, 12pm, 3pm, 6:30pm |
| 16:8 IF (noon start) | lunch, snack, dinner, evening | 12pm, 3pm, 6:30pm, 8:30pm |
| 20:4 IF | meal-1, meal-2 | 4pm, 7pm |
| OMAD | dinner | 6pm |
| Morning faster + pre-bed | lunch, snack, dinner, pre-bed | 12pm, 3pm, 6:30pm, 9:30pm |
Calorie distribution should match the window: for 4-meal IF, roughly 30% / 12% / 35% / 23% across slots (larger meals at lunch and dinner, lighter snacks). Adjust to keep protein per-meal at 0.3–0.4 g/kg (~25–50g) for optimal MPS distribution.
If body fat % is known, use Katch-McArdle instead of Mifflin-St Jeor — it's more accurate for lean or obese individuals.
Mifflin-St Jeor BMR (validated as most accurate predictive equation for the general population, ±10% for most adults):
Men: BMR = (10 × weight_kg) + (6.25 × height_cm) − (5 × age) + 5
Women: BMR = (10 × weight_kg) + (6.25 × height_cm) − (5 × age) − 161
Katch-McArdle (use if body fat % is known — more accurate at body-composition extremes):
BMR = 370 + (21.6 × lean_mass_kg) where lean_mass_kg = weight_kg × (1 − bodyfat%)
TDEE = BMR × activity multiplier:
| Level | Multiplier | Description |
|---|---|---|
| Sedentary | 1.2 | Desk job, minimal deliberate exercise |
| Lightly active | 1.375 | 1–3 sessions/week |
| Moderately active | 1.55 | 3–5 sessions/week |
| Very active | 1.725 | 6–7 sessions/week |
| Extra active | 1.9 | Athlete or physical labor + training |
People consistently overestimate their activity — when in doubt, pick the lower multiplier and adjust upward after 2 weeks of real-world data.
Goal adjustment:
Fat loss: TDEE − 20–25% (typically 400–600 kcal deficit). Targets ~0.5–1% bodyweight/week. Steeper deficits accelerate muscle loss.
Maintenance: TDEE ± 0
Muscle gain: TDEE + 10–15% (typically 200–400 kcal surplus). Larger surpluses mostly add fat, not muscle, in trained individuals.
| Goal | Target | Evidence |
|---|---|---|
| General health / sedentary | 0.8–1.2 g/kg | RDA is 0.8 g/kg — the minimum to prevent deficiency, not an optimum |
| Muscle gain + resistance training | 1.6–2.2 g/kg | Morton et al. 2018 meta-analysis (49 RCTs, n=1,863, BJSM): gains in fat-free mass plateau at 1.62 g/kg/day (95% CI: 1.03–2.20). The upper CI bound (~2.2 g/kg) is recommended for those maximizing hypertrophy. Confirmed by Nunes et al. 2022 (74 RCTs). |
| Fat loss (preserving lean mass) | 1.8–2.7 g/kg | Higher end of range compensates for the muscle-sparing effect of protein during energy deficit |
| Endurance athletes | 1.2–1.6 g/kg | ISSN position stand |
In imperial: 1.6–2.2 g/kg ≈ 0.7–1.0 g/lb. Distribute across 3–5 meals at 0.3–0.4 g/kg per meal (~25–40 g) — muscle protein synthesis response plateaus per-sitting.
Minimum ~0.5 g/kg bodyweight (hormone synthesis floor). Typical range 0.8–1.2 g/kg, or 20–35% of calories. Going below 20% long-term risks fat-soluble vitamin and hormone issues.
Fill remaining calories after protein and fat are set. carbs_g = (target_kcal − protein_g×4 − fat_g×9) / 4
14 g per 1,000 kcal (Dietary Guidelines for Americans) — roughly 25 g/day for women, 38 g/day for men at maintenance. Most plans undershoot this. Check it explicitly.
Adherence-first meal design — build meals around foods the user already eats and enjoys. The goal is to fit their existing food culture into their macro targets, not to impose a "clean eating" template.
Use their stated preferences — if they said they like burritos, build a macro-friendly burrito bowl. Don't replace it with a salad.
Flavor over restriction — use real cheese, butter, sauces in controlled portions rather than eliminating them. A plan that tastes good gets followed.
Practical portions — specify gram weights for precision but also include household measures (e.g., "200g / ~1 cup cooked") so people without scales can approximate.
Protein anchoring — every meal should have a clear protein source as its anchor. Build the rest of the plate around it.
Vegetable integration — don't just add a side salad. Integrate vegetables into the meal itself (peppers in a stir-fry, spinach in a smoothie, roasted broccoli with garlic butter). This is especially important for users who say they don't like vegetables.
Rotate 3–4 templates per meal slot across the week — enough variety to prevent burnout, enough repetition to keep prep and shopping simple. The sweet spot:
2–3 lunch templates — repeat 2–3 times each across the week
5–6 dinner templates — one unique dinner most nights (this is where people want variety)
2–3 snack templates — these can repeat heavily; people don't mind eating the same snack daily
1–2 evening/pre-bed options — simple, protein-focused, low effort
## Monday — Target: 2,400 kcal | 186P / 225C / 84F (16:8 IF, noon start)
Lunch (710 kcal): Chicken burrito bowl — chicken thigh 170g + rice 200g + black beans 80g + salsa + cheese 25g + avocado 40g → 48P / 78C / 22F
Snack (310 kcal): Greek yogurt 200g + honey 10g + almonds 20g → 24P / 22C / 14F
Dinner (710 kcal): Garlic butter steak & potatoes — sirloin 200g + baby potatoes 250g + roasted broccoli 120g + butter 10g → 58P / 52C / 30F
Evening (630 kcal): Protein shake — whey 35g + banana + PB 20g + milk 250ml + oats 30g → 48P / 62C / 20F
Daily total: 2,360 kcal | 178P / 214C / 86F | Fiber: ~21g
Nutrition data sources — for accurate macros, query the USDA FoodData Central API rather than guessing:
Base URL: https://api.nal.usda.gov/fdc/v1/ — free API key from api.data.gov, 1,000 req/hr limit
Search: GET /foods/search?query=chicken breast&api_key=KEY
Lookup: GET /food/{fdcId}?api_key=KEY returns full nutrient profile per 100g
Python clients on PyPI: fooddatacentral (simple), usda-fdc (includes DRI comparison + recipe aggregation)
Data is public domain (CC0). Branded foods update monthly.
Aggregate ingredients across all days, round to purchasable units, organize by store section:
PROTEINS: Chicken thighs 800g · Chicken breast 700g · Sirloin steak 400g · Salmon 180g · Eggs 1 dozen
DAIRY: Greek yogurt 550g · Cottage cheese 450g · Cheddar cheese 105g · Whole milk 750ml
PRODUCE: Bananas 4 · Bell peppers 4 · Broccoli 1 head · Avocado 2 · Baby potatoes 250g
GRAINS/PANTRY: White rice 1kg · Flour tortillas 8 · Oats 90g · Black beans 1 can
NUTS/OILS: Peanut butter 1 jar · Almonds 20g · Olive oil · Honey
Aggregation rules:
Combine same ingredients across days (e.g., 3 days × 200g rice = 600g → round to 1kg bag)
Use purchasable units (1 can, 1 bunch, 1 dozen) not fractional quantities
Group by store section for efficient shopping flow
Note items that can be bought frozen for longer shelf life
Single prep session (Sun, ~90 min): Cook all grains. Roast two sheet-pans of proteins + veg. Portion into containers. Prep overnight oats for 3 days.
USDA-backed refrigerated shelf life: cooked poultry/meat 3–4 days · cooked fish 3–4 days · cooked grains 4–6 days · cut raw veg 3–5 days. Freeze anything for day 5+.
Prep guide format: Number each step with estimated time (active + passive). Group by what can run in parallel (e.g., "while rice cooks, grill chicken"). The prep guide should feel like a cooking flow, not a to-do list.
General target: ~35 ml per kg bodyweight per day (e.g., 93 kg × 35 = 3.25 L). Increase by 500–750 ml on training days. Signs of adequate hydration: pale yellow urine by mid-morning.
Practical tips: keep a water bottle at the desk, drink 500 ml upon waking, have 250 ml with each meal.
Only recommend supplements with strong evidence. Do not recommend proprietary blends or unproven supplements.
| Supplement | Dose | Evidence | Who needs it |
|---|---|---|---|
| Creatine monohydrate | 3–5 g/day, no loading needed | Most-studied supplement in sports nutrition. Increases strength, lean mass, and exercise capacity (ISSN position stand, 2017). | Anyone doing resistance training |
| Vitamin D3 | 1,000–2,000 IU/day | Widespread deficiency, especially at higher latitudes. Supports bone health, immune function, and may support muscle function. | Most people, especially if limited sun exposure |
| Omega-3 (EPA/DHA) | 1–2 g combined EPA+DHA/day | Anti-inflammatory, cardiovascular benefits. Only if fish intake is <2 servings/week. | People who don't eat fish regularly |
| Magnesium | 200–400 mg/day (glycinate or citrate) | Common deficiency. Supports sleep, muscle recovery, stress. | If dietary intake is low (common on calorie deficits) |
Do NOT recommend: testosterone boosters, fat burners, BCAAs (redundant if protein is adequate), collagen for muscle building, "detox" products.
Plans are not static. Build in adjustment checkpoints:
| Trigger | Action |
|---|---|
| After 2 weeks | Compare average weekly weight loss rate to target. If losing >1.5% BW/week, increase calories by 100–200. If weight is flat, decrease by 100–200 or verify tracking accuracy first. |
| Every 5 kg (~11 lbs) of weight change | Recalculate BMR and TDEE — they decrease as body mass drops. A 10 lb loss typically reduces TDEE by ~100–150 kcal. |
| Plateau (>3 weeks no change in 7-day rolling average) | First check: are portions actually being measured? Calorie creep from eyeballing is the #1 cause. Then: add 1–2 cardio sessions/week OR reduce calories by 100–150. Do NOT slash calories aggressively. |
| Goal reached | Reverse diet: increase calories by 100/week back toward maintenance over 4–6 weeks. Do not jump straight to TDEE — leptin and metabolic adaptation need time. |
| Taste fatigue / boredom | Swap out stale meals using alternatives from the same slot. Keep the same macros — change the food, not the structure. |
Daily weight fluctuates ±1–2 kg (2–4 lbs) from water, glycogen, gut contents, sodium intake, and menstrual cycle. This is normal and not fat gain.
Weigh daily, same time (after waking, after bathroom, before eating)
Use a 7-day rolling average — this is your actual trend line
Compare weekly averages, not individual days
Expect a 2–4 lb drop in week 1 that's mostly water/glycogen, not fat. Real fat loss rate starts from week 2–3.
Beginner (<6 mo): Full body 3×/week (Mon/Wed/Fri). Each session: 1 squat pattern, 1 hinge, 1 push, 1 pull, 1 carry/core. 3×8–12. Simplest progression: add 2.5 kg when all sets hit the top of the rep range.
Intermediate (6 mo–2 yr): Upper/Lower 4×/week. Mon upper-push emphasis, Tue lower-quad, Thu upper-pull, Fri lower-hinge.
Advanced: Push/Pull/Legs 5–6×/week.
Volume guidelines (per muscle group per week, meta-analytic consensus): 10–20 hard sets for hypertrophy. Below 10 is maintenance; above 20 shows diminishing returns and rising injury risk for most.
Rep ranges: 3–6 strength-biased · 6–15 hypertrophy (all ranges build muscle if taken near failure; hypertrophy is not rep-range-specific) · 15+ endurance-biased.
Progressive overload is mandatory — log every session. No log → no plan.
Recovery: ≥1 full rest day/week. 7–9 hrs sleep. Deload (−40–50% volume) every 4–6 weeks.
Cardio: General health → 150 min/week moderate (WHO guideline). Fat-loss phase → add 2–3 × 20–30 min sessions, steady-state or intervals. Muscle-gain phase → keep cardio to 1–2 light sessions to minimize interference.
Every meal plan MUST be delivered as an interactive React + Vite web app. Do not output plans as plain text or markdown — always build and deploy a visual website.
Most people use a meal planner in the kitchen (cooking), at the grocery store (shopping), or in bed (planning). Design for a phone screen first, then scale up.
Single-column layout on mobile — no multi-column calendar grids that require horizontal scrolling
Day selector as a horizontal scrollable pill bar at the top (not a 7-column grid)
Meal cards are full-width, showing meal name, time, calories, and protein at a glance — tap to expand for full detail
Bottom navigation for section switching (Plan / Shop / Prep)
Sticky headers for context while scrolling
Progress card (weight goal, deficit, timeline)
Day selector (horizontal pills showing day name + total kcal)
Macro rings/charts for selected day (protein, carbs, fat vs targets)
Meal cards for selected day (compact: name, time, kcal, protein — tap to drill in)
Weekly averages card at the bottom
Macro summary bar (4-up: kcal, P, C, F with targets)
Expanded meal cards showing all food items with portions
Per-meal macro breakdown (color-coded: protein=green, carbs=amber, fat=blue)
Checkboxes for each item (state persists during session)
Progress counter (checked/total)
Organized by store section (Proteins, Dairy, Produce, Grains/Pantry, Nuts/Oils)
Hero card explaining the prep concept and total time
Numbered steps with time estimates and descriptions
Organized as a cooking flow (what to do while waiting for passive cook times)
Meal swaps — for each meal slot, offer 2–3 alternatives the user can click to swap in. Show a small "swap" icon on each meal card that reveals alternatives in a dropdown or modal.
Rotation plans — if the user wants variety (e.g., "don't repeat the same dinner twice in 2 weeks"), build a 2-week rotation with different week tabs.
High/low day cycling — for carb cycling or calorie cycling plans, label each day (e.g., "High Day — 2,400 kcal" vs "Low Day — 1,800 kcal") and color-code the day header accordingly.
Clean, appetizing design — health-focused color palette (greens, warm neutrals), rounded cards, readable portions
Mobile-first — single column on mobile, expanding gracefully on desktop
Tap to expand — meal cards expand on tap to show full ingredient list and nutrition detail
Bottom navigation — persistent bottom nav bar for Plan / Shop / Prep (hidden on detail/drill-in pages)
Sticky headers — keep context (day name, section title) visible while scrolling
Embed all plan data as JSON in the app (no backend needed):
interface MealPlan {
weeks: Week[];
profile: {
goal: string;
currentWeight: number;
targetWeight: number;
bmr: number;
tdee: number;
target_kcal: number;
deficit: number;
macros: Macros;
projectedWeeks: number;
};
shoppingList: ShoppingSection[];
prepGuide: PrepStep[];
}
interface Week {
label: string;
days: Day[];
}
interface Day {
name: string;
label?: string; // "High Day", "Rest Day", "Meal Prep Day", etc.
target_kcal: number;
meals: Meal[];
}
interface Meal {
slot: string; // dynamic: "lunch", "snack", "dinner", "evening", "meal-1", etc.
name: string;
items: { food: string; portion: string }[];
macros: Macros;
alternatives?: Meal[];
}
interface Macros {
protein: number;
carbs: number;
fat: number;
fiber: number;
kcal: number;
}
interface ShoppingSection {
section: string;
items: { item: string; qty: string }[];
}
interface PrepStep {
step: string;
time: string;
description: string;
}
Adherence beats optimization — a 90%-adhered B+ plan beats a 50%-adhered A+ plan. Build around foods the user already likes.
Protein is the anchor macro — set it in g/kg first, then fill carbs/fat by preference.
Budget a 10% flex buffer — plans that forbid all unplanned food get abandoned. If someone goes 200 kcal over one day, the plan should still work over the week.
Re-calculate TDEE after weight changes ~5 kg — BMR shifts with body mass.
Track for 2 weeks before adjusting — daily weight fluctuates ±1–2 kg from water/glycogen/gut contents. Use a 7-day rolling average.
Respect the eating window — never schedule meals outside the user's stated eating window. If they fast until noon, the first meal is lunch, not "late breakfast."
Make vegetables invisible — for users who don't love vegetables, integrate them into meals (spinach in smoothies, peppers in stir-fries, zucchini in pasta sauce) rather than serving them as plain sides.
Protein in every meal — every single meal/snack should have a meaningful protein source (≥20g). No meal should be carbs + fat only.
Evening meals matter — casein protein, cottage cheese, Greek yogurt, and protein puddings are ideal pre-bed options. They're slow-digesting and satisfy sweet cravings.
Nutritional values are estimates (±10–15% even with USDA data — portion eyeballing adds more error)
TDEE formulas are population averages with ~10% individual error — real-world tracking over 2–3 weeks is the only way to find someone's true maintenance
Not a substitute for a registered dietitian, especially for medical conditions, disordered eating history, or pregnancy
Training templates are generic — modify around injuries and individual response
Hydration needs vary significantly with climate, altitude, and individual physiology
Supplement recommendations are general — individual needs vary based on bloodwork and existing diet