We raised a $50M Series B led by BenchmarkLearn more
Alma logo — MCP server on Gumloop

Alma

Track your nutrition and meals with your Alma AI coach.

Book a demo

Installation

Set up the Alma MCP server in Gumloop

Do this once to provision your hosted server URL.

1

Create a Gumloop account

To use this MCP, you need a Gumloop account. If you don't have one yet, sign up and start a 14-day free trial.

2

Add and authorize the Alma server

In Gumloop, open Connectors and add Alma. Depending on the app, you'll either sign in via OAuth or paste an API key. Either way, the credential is stored securely in Gumloop.

Then use it in your client

Use in GumloopUse in Gumloop
1

Use Alma in an agent

Once Alma is set up, just open any Gumloop agent, add Alma as a connector, and start chatting with the agent.

Tools (10)

  • Get User Profile

    User context: name, email, timezone, activity, gender, height/weight, and active goal descriptions. Call early when personalizing advice.

  • Get Meal Data

    Logged meals and nutrition for the user. Default detail_level is 'detailed': each day includes meal_details with foods and their calories, protein, carbs, and fat. Use the top-level micronutrient_data_available flag before answering micronutrient questions; when it is false, the data is not available. Use detail_level='summary' only for daily calorie/macro totals without foods. Pass start_date and end_date (YYYY-MM-DD) together for a fixed range, else use days. days must be 1-365.

  • Get Alma Score Data

    Alma Score (0-100) and 10 diet-quality categories (veg, fruit, whole grains, nuts/legumes, omega-3s, healthy fats, processed meat, sugar, sodium, alcohol). Use detail_level='detailed' for per-day category breakdown; 'summary' for rollups only. Same date rules as get_meal_data (days, or start_date+end_date). days must be 1-365.

  • Get Goals

    Active nutrition/fitness goals with targets and progress (up to 10). No parameters. Use with get_meal_data to compare intake to targets.

  • Get Health Metric History

    Synced Apple Health and Journey metrics, including smart-scale body composition (body fat, lean mass, BMI), sleep, HRV, heart rate, steps, active energy, blood glucose, and other authorized health signals. Returns latest values, daily history, and trends versus the preceding period. Use get_weight_history separately for body weight and get_workout_history for individual workouts. Missing rows mean unknown or not synced, never zero. Same date rules as get_meal_data.

  • Get Workout History

    Workout history and exercise minutes from synced HealthKit/Journey data. Returns individual workouts, workout-duration totals, calories burned, common activities, and daily exercise_minutes when available. Same date rules as get_meal_data (days, or start_date+end_date). days must be 1-365.

  • Get Weight History

    Weight entries from HealthKit-linked data: recent values, trend (gaining/losing/maintaining), and change over the lookback. Default days=30. days must be 1-365.

  • Get Pantry Items

    Pantry contents for the authenticated user (own + shared via a family circle), grouped by category. Empty pantry returns an empty payload with has_data=false.

  • Get Grocery List

    Grocery list for the authenticated user (own + shared via a family circle). include_completed=true (default) returns purchased items too. Completed items are sorted to the bottom.

  • Log Food Items

    Add one confirmed meal containing 1-20 fully specified food items to the authenticated user's Alma log. Call only after the user explicitly asks to log the food and confirms the item names, portions, calories, protein, carbs, and fat. This writes user-owned estimated snapshots, not shared catalog foods. client_operation_id must be a new UUID and must be reused unchanged when retrying the same request. consumed_at must include a UTC offset and be within the last 365 days; omit it to log now. Never use this tool for tentative suggestions or meal plans.

Ship Alma agents in minutes

Connect any AI agent to 100+ MCP servers, zero setup.
Book a demo
Gradient