Nutrition MCP
MCP serverMonitoring & opsThis integration turns your AI into a personal nutrition tracker. Once added, it can log the meals you eat, keep an eye on your macros, and show you what you ate on previous days. If you already track food in another app, you can bring that history over so nothing is lost.
Use Nutrition MCP in Claude, ChatGPT or Ahel Desktop
Free. Sign in, add Nutrition MCP and connect your AI. About a minute.
Also: Claude Code · Cursor · Codex
Then ask your AI: use Nutrition MCP
Needs your own Nutrition MCP account. You sign in to it and approve access when you connect.
Details
Available today. Use it from your connected AI after setup.
Add the integration, then tell your AI about a meal to log it, or ask it to import your history from the app you currently use.
What your AI can do with it
- Log meals as you eat them
- Track your macros over time
- Review your eating history from past days
- Import your existing food log from another app
Install Nutrition MCP
The server’s own address, for the clients that take one directly. Or connect Ahel once and every client you use reads it from one address, with the account kept on Ahel rather than in each client’s config.
Claude Code
claude mcp add --transport http --scope user nutrition-mcp 'https://nutrition-mcp.com/mcp'Run it once in your project, then open /mcp to approve any sign-in the server asks for.
Claude Desktop
https://nutrition-mcp.com/mcpAdd a custom connector in Settings, paste this address, and approve the sign-in.
Cursor
cursor://anysphere.cursor-deeplink/mcp/install?name=nutrition-mcp&config=eyJ1cmwiOiJodHRwczovL251dHJpdGlvbi1tY3AuY29tL21jcCJ9Open the link and Cursor adds the server at that address.
ChatGPT
https://nutrition-mcp.com/mcpIn Settings, enable Developer mode, create an MCP app, and paste this address. Your plan and workspace must allow custom apps.
Codex
codex mcp add nutrition-mcp --url 'https://nutrition-mcp.com/mcp'Run it once, then sign in with codex mcp login nutrition-mcp if the server asks for an account.
From the project's README
As published by akutishevsky/nutrition-mcp in README.md.
A remote MCP server for personal nutrition tracking — log meals with calories, macros, fiber, total and added sugar, saturated and trans fat and caffeine (optionally ingredient by ingredient), look up generic foods in USDA FoodData Central and packaged products by barcode in Open Food Facts, record where each nutrient value came from, save meals you eat often and log them again by name, log water, body weight and body measurements, review nutrition history, and import an existing food diary from another app, all through conversation. Alcohol tracking is opt-in and off by default.
Help me pay for the servers on Patreon
Table of Contents
- Quick Start
- Demo
- Tech Stack
- MCP Tools
- MCP Resources
- Self-hosting
- 0. Get the code
- 1. Supabase setup
- 2. Environment variables
- 3. Google sign-in (optional)
- Development
- Testing and quality
- Connect to Claude.ai
- Apple Health sync
- Troubleshooting
- API Endpoints
- Deploy
- Support & security
- Data sources
- License
Quick Start
Already hosted and ready to use — just connect it to your MCP client:
https://nutrition-mcp.com/mcp
On Claude.ai: open Nutrition MCP in the Claude directory and click Connect (see Connect to Claude.ai below)
On first connect, sign in with Google or enter an email and password and choose Create account. Your data persists across reconnections. Trouble connecting? See Troubleshooting.
By connecting you agree to the Terms of Service; how your data is handled is in the Privacy Policy.
Switching from another tracker? See the nutrition-app alternatives — how it compares to MyFitnessPal, Cronometer, Lose It!, MacroFactor, Yazio, and Lifesum. Bring your history with you: say "import my meals" and an importer opens in the chat, where you pick the CSV you exported from your old app, map its columns, and check what will be added before anything is saved. Exports from MyFitnessPal, Cronometer, Lose It! and MacroFactor are recognised automatically; any other CSV works by mapping its columns yourself. In clients that can't show in-chat panels, paste the export instead and the AI imports it for you. If your export has an alcohol column and you want it kept, turn alcohol tracking on before importing — the importer skips that column while tracking is off, and re-importing the same file later won't backfill it.
Demo
Read the story behind it: How I Replaced MyFitnessPal and Other Apps with a Single MCP Server
Tech Stack
- Bun — runtime and package manager
- Hono — HTTP framework
- MCP SDK — Model Context Protocol over Streamable HTTP
- Supabase — PostgreSQL database + user authentication
- OAuth 2.0 — authentication for Claude.ai and other MCP clients: per-client dynamic client registration (RFC 7591) with exact redirect-URI matching, mandatory PKCE (
S256), token-endpoint client authentication (none,client_secret_post,client_secret_basic), codes and refresh tokens bound to the client they were issued to,isson every authorization response (RFC 9207), and client secrets, access tokens, refresh tokens and codes stored only as a SHA-256 hash. Access tokens last 24 hours; refresh tokens last 90 days from their last use and rotate on every refresh; expired tokens and never-used client registrations are swept hourly
MCP Tools
| Tool | Description |
|---|---|
log_meal | Log a meal with description, type, calories, macros, fiber, total and added sugar, saturated and trans fat, alcohol, caffeine (mg), notes — from text or a photo of your plate; each value records its source |
start_meal_import | Open the in-chat CSV importer: pick an export from another app, map its columns, preview, confirm |
bulk_import_meals | Write up to 50 imported rows per call — each row validated, duplicates skipped so a re-send is safe while the timezone is unchanged |
lookup_barcode | Look up a packaged product's label nutrition by barcode via Open Food Facts, including added sugar when listed (read from a photo or typed; data © OFF contributors, ODbL) |
search_foods | Search USDA FoodData Central generic foods by English name; up to 10 candidates with their ids and macros per 100 g |
get_food_macros | USDA FoodData Central values for one generic food by its id, per 100 g or scaled to amount_g, with its listed portion sizes |
get_meals_today | Get all meals logged today, one compact line per meal with its id; detail: "full" adds notes |
get_meals_by_date | Get meals for a specific date (YYYY-MM-DD), one compact line per meal; detail: "full" adds notes |
get_meals_by_date_range | Get meals between two dates (inclusive), up to 31 days per call; one compact line per meal, detail: "full" adds notes |
search_meals | Search past meals by keyword, grouped into recurring variations (counts, last logged, typical macros) |
get_nutrition_summary | Daily nutrition totals + goal progress for a date range, up to 92 days per call |
update_meal | Update any fields of an existing meal |
delete_meal | Delete a meal by ID |
save_meal | Save a meal under a name, with values per serving and optional ingredients, to log again later |
log_saved_meal | Log a saved meal, scaled by servings, with single ingredients set to the amount eaten or left out |
get_saved_meals | List saved meals with their values and ingredients, optionally filtered by name |
update_saved_meal | Change a saved meal's name, values or ingredients; meals already logged from it keep theirs |
delete_saved_meal | Delete a saved meal; meals already logged from it keep their values |
set_nutrition_goals | Set daily calorie, macro, fiber and water targets to reach, total-sugar/added-sugar/alcohol/caffeine limits to stay under, plus an optional target weight |
get_nutrition_goals | Get the current daily targets and limits |
get_goal_progress | Get intake vs. targets and limits for a given day (default: today), plus latest weight vs. target |
log_water | Log a hydration entry in milliliters |
get_water_today | Get today's water intake total and entries |
get_water_by_date | Get water intake for a specific date |
delete_water | Delete a water log entry by ID |
log_weight | Log a body-weight measurement in kg or lb (converted and stored server-side) |
get_weight_today | Get today's weight entries |
get_weight_by_date | Get weight entries for a specific date |
get_weight_by_date_range | Get weight entries between two dates (inclusive), grouped by day, up to 366 days per call |
get_weight_trends | Weight trend: smoothed trend weight, weekly rate, latest, overall change, min/max, goal progress; widget zooms to 90 days, 1 year or all history |
update_weight | Update an existing weight entry |
delete_weight | Delete a weight entry by ID |
set_weight_unit | Set the preferred weight unit (kg or lb; null to clear) |
log_body_measurement | Log a tape measurement (waist, hips, neck, chest, shoulders, upper arm, forearm, thigh, calf) in cm or in |
get_body_measurements | List body measurements by day, optionally for one site; last 30 days by default, up to 366 days per call |
update_body_measurement | Update a measurement's value, unit, time or notes (the site is fixed) |
delete_body_measurement | Delete a body measurement by ID |
set_length_unit | Set the preferred length unit (cm or in; null to clear) |
get_trends | 7/14/30-day averages, std dev, streaks, best/worst day; group_by week/month/quarter/year averages per logged day vs targets then in effect |
get_meal_patterns | Pre-aggregated behavioural patterns (breakfast effect, late dinner, weekend vs weekday, outliers) |
export_all_data | Export everything stored about you — logs, saved meals and ingredients, where each nutrient value came from, goals, profile, account, telemetry, app connections — as one ZIP of CSVs + README; 60-minute link |
get_profile | Get timezone (+ local date/time), widget language, weight and length units, widget display and alcohol tracking in one call |
set_timezone | Set the user's IANA timezone (e.g. America/Los_Angeles) |
set_language | Set the UI language for in-chat widgets (dashboards, charts) — not the language the AI replies in |
get_current_time | Get the current date and time in the user's timezone, plus the UTC instant — for hosts with no clock in context |
set_widget_display | Enable or disable the in-chat visual widgets (dashboards, rings, charts); enabled by default |
set_alcohol_tracking | Turn alcohol tracking on or off (off by default) and choose US standard drinks or UK units; turning it off hides alcohol rather than deleting it |
delete_account | Permanently delete the user's Nutrition MCP account and all data it stores about them |
MCP Resources
| URI | Description |
|---|---|
nutrition://weekly-summary | Rolling 7-day digest (averages vs targets, best/roughest day) for proactive pulls |
Self-hosting
0. Get the code
git clone https://github.com/akutishevsky/nutrition-mcp.git
cd nutrition-mcp
bun install
cp .env.example .env # fill in real values as you go through the steps below
Requires Bun 1.x (matches the Dockerfile's oven/bun:1 base image; no exact minor version is pinned).
Making it yours: The public site includes the maintainer's personal bits — Google Analytics, Microsoft Clarity, Patreon/GitHub/contact links, and the
nutrition-mcp.comdomain. Runbun run gen:allto produce the public pages, thenbun run depersonalizeto strip the personal bits in one pass (the consent-gated analytics snippet, the consent banner and the footer "Cookie settings" button, plus the analytics CSP hosts, the Support/Contact sections, social links, and the domain → ayour-domain.complaceholder). Usebun run depersonalize --dryto preview without writing. Afterwards, swap in your ownpublic/og.png,favicon.ico,favicon.svg,apple-touch-icon.png,icon-192.pngandicon-512.png, replace the domain placeholder with your real domain, and rewrite the analytics paragraphs of the privacy policy and terms (src/copy/legal*.ts), which still name both services. Its edits to the generated pages are undone bybun run gen:all— to regenerate, first change the sources it reads from (inscripts/site-partials.ts, either your ownGA_MEASUREMENT_ID/CLARITY_PROJECT_ID, whichanalyticsHead()loads only after a visitor accepts the consent banner, orANALYTICS_ENABLEDset to false to drop the snippet, banner and button altogether;SITEinsrc/routes.ts), then run it again. This script only touches the generated files underpublic/, the import widget's support address,src/index.tsand the security.txt contact insrc/security-txt.ts(blanked, so/.well-known/security.txtanswers 404 until you set your own) — it doesn't touch this README orSECURITY.md, so if you're publishing your own fork, rewriteSECURITY.mdfor your deployment and also edit or remove the Patreon line near the top of this file and the Medium link in Demo.
1. Supabase setup
-
Create a Supabase project.
-
Enable Email Auth (Authentication → Providers → Email) and disable email confirmation.
-
Apply the schema. The full schema lives in
supabase/migrations/. With the Supabase CLI:supabase link --project-ref <your-project-ref> supabase db pushThis creates every table, index, RLS policy, and foreign key the app needs. No local Postgres is involved — migrations run against your hosted project.
-
Copy the service role key from Project Settings → API and use it as
SUPABASE_SECRET_KEY.
2. Environment variables
Shortened here. Read the whole README on GitHub.
Signals
- GitHub stars
- 70
- Forks
- 28
- Last commit
- Oct 2026
Advanced
- Delivery
- nutrition-mcp MCP server → your Ahel connector (mcp.ahel.ai) → your AI.
- Item type
- mcp-server
- Key
io-github-akutishevsky-nutrition-mcp- Source
- github.com/akutishevsky/nutrition-mcp
- Hosted endpoint
https://nutrition-mcp.com/mcp
github.com/akutishevsky/nutrition-mcp
More in Monitoring & ops
MCP server
More in Monitoring & opsWorld Monitor
MCP server · koala73
More in Monitoring & opsNetdata
MCP server · netdata
More in Monitoring & opsotakit
MCP server · otakit
More in Monitoring & opsStatuser
MCP server · statuser-cloud
More in Monitoring & opsaltmetric-mcp
MCP server · altmetric
More in Monitoring & ops