How to Build a Personalized Fitness Plan Generator in FlutterFlow

Create a new page called FitnessAssessmentPage. Add a ScrollView containing a Column with the following input widgets: a Slider for age (16–75), two TextFields for height (cm) and weight (kg), a DropdownButton for primary goal (options: Lose Weight, Build Muscle, Improve Endurance, Maintain Fitness), a Multi-select chip widget listing available equipment (Bodyweight Only, Dumbbells, Barbell, Resistance Bands, Full Gym), and a ToggleButtons row for days per week (3, 4, 5, or 6). Bind each widget to a Page State variable of the appropriate type. Add a prominent medical disclaimer Text widget at the bottom of the form in red: 'Consult your doctor before starting any new exercise program, especially if you have a medical condition, injury, or are new to exercise.' Add a Checkbox that must be ticked to enable the 'Generate My Plan' button.

In your Firebase project, create a Cloud Function called generateFitnessPlan. It accepts a JSON body with fields: age, heightCm, weightKg, goal, equipment (array), daysPerWeek, and optionally previousPlanSummary and recentLogs for progressive regeneration. Build a detailed system prompt that positions the AI as a certified personal trainer. Construct a user message summarising the assessment data. Call your AI API (OpenAI or Anthropic) with the system prompt and user message, requesting a structured JSON response with an array of day objects, each containing dayName, focus, and an exercises array with name, sets, reps, rest, and notes. Return the parsed JSON to the Flutter client. Store your API key using Firebase Functions config or Secret Manager — never hardcode it.

1// functions/index.js
2const functions = require('firebase-functions');
3const admin = require('firebase-admin');
4const axios = require('axios');
5admin.initializeApp();
6
7exports.generateFitnessPlan = functions.https.onCall(async (data, context) => {
8  if (!context.auth) throw new functions.https.HttpsError('unauthenticated', 'Login required');
9
10  const { age, heightCm, weightKg, goal, equipment, daysPerWeek, previousPlanSummary } = data;
11  const apiKey = process.env.OPENAI_API_KEY;
12
13  const systemPrompt = `You are a certified personal trainer with 10 years of experience.
14Create safe, progressive workout plans tailored to the individual.
15Always prioritize proper form and injury prevention.
16Return ONLY valid JSON matching this schema:
17{"days": [{"dayName": string, "focus": string, "exercises": [{"name": string, "sets": number, "reps": string, "rest": string, "notes": string}]}]}`;
18
19  const userMessage = `Create a ${daysPerWeek}-day/week workout plan for:
20- Age: ${age}, Height: ${heightCm}cm, Weight: ${weightKg}kg
21- Goal: ${goal}
22- Available equipment: ${equipment.join(', ')}
23${previousPlanSummary ? '- Previous plan: ' + previousPlanSummary + ' (increase intensity by 5-10%)' : ''}`;
24
25  const response = await axios.post('https://api.openai.com/v1/chat/completions', {
26    model: 'gpt-4o',
27    messages: [{ role: 'system', content: systemPrompt }, { role: 'user', content: userMessage }],
28    response_format: { type: 'json_object' },
29  }, { headers: { Authorization: `Bearer ${apiKey}` } });
30
31  const plan = JSON.parse(response.data.choices[0].message.content);
32  return plan;
33});

Create a Custom Action called callGenerateFitnessPlan. It reads all Page State variables from the assessment form and calls FirebaseFunctions.instance.httpsCallable('generateFitnessPlan') with them as a map. On success, save the returned plan JSON to a new Firestore document in the fitness_plans collection with the current user's UID, a createdAt timestamp, and a planVersion integer starting at 1. Also save the plan JSON to an App State variable called currentPlan so you can display it immediately without a Firestore read. Navigate to the PlanViewPage. Wire this action to the 'Generate My Plan' button. Show a full-screen loading overlay with an animated logo while the AI generates the plan — this typically takes 3–8 seconds.

Create PlanViewPage. Retrieve the currentPlan from App State. The plan's days array length determines the number of tabs — use a Custom Widget or FlutterFlow's Tab Bar widget with dynamic tab count. Each tab label is the dayName field (e.g., 'Monday - Push', 'Wednesday - Pull'). Inside each tab, add a Text widget for the focus description, then a ListView of exercise cards. Each card shows the exercise name in a bold Text, and a Row with chips for sets x reps, rest time, and a small info icon that opens a Popup showing the notes field. Add a 'Mark All Complete' button at the bottom of each tab that logs the day's workout to Firestore.

When the user taps 'Mark All Complete' on a day tab, create a Firestore document in the workout_logs collection with fields: userId, planId, dayName, completedAt, and exercisesCompleted (the full exercise list). After logging, check if all days in the current plan have been completed within the last 7 days. If so, show a Congratulations Dialog with two buttons: 'Keep This Plan' and 'Generate Progressive Plan'. The 'Generate Progressive Plan' button calls callGenerateFitnessPlan again but this time passes previousPlanSummary — a string summarising the completed plan's exercises and volumes. The Cloud Function uses this to increase intensity by 5–10% in the new plan. Increment planVersion in Firestore so users can browse their plan history.