All-In-One: FoodFinder + LoopInsights + AutoPresets

[taylorpatterson-T1D]

All-In-One Feature Branch: FoodFinder + LoopInsights + AutoPresets

This branch bundles all three features into a single build-ready LoopWorkspace. Clone it, open in Xcode, build, and you get everything.

Included Features

1. FoodFinder

AI-powered food identification and nutritional analysis. Scan barcodes, use your camera, voice search, or type to find foods and auto-populate carb entries. Supports OpenAI, Anthropic, Google, and custom endpoints.

• PR: AutoPresets: Automatically activate your Loop override presets during walking/running activity LoopKit/Loop#2402

2. LoopInsights

AI-powered therapy settings advisor. Analyzes your glucose, insulin, and carb data to suggest adjustments to Carb Ratio, ISF, and Basal Rate schedules with clinical reasoning. Includes AGP charts, trends dashboard, AI chat, and suggestion history with one-tap revert.

• PR: LoopInsights: AI-powered therapy settings analysis LoopKit/Loop#2405

3. AutoPresets

Automatically apply Loop presets based on motion and activity data from Apple Health. Set rules to trigger presets during exercise, sleep, or other activities.

• PR: FoodFinder: AI-powered & barcode food identification for carb entry LoopKit/Loop#2404

---

How to Build with Xcode

Step 1: Clone the repository

Open Terminal and paste this entire block as one command:

BUILD_DIR=~/Downloads/BuildLoop/Loop-$(date +%Y-%m-%d-%H%M%S) && mkdir -p "$BUILD_DIR" && cd "$BUILD_DIR" && git clone --branch feat/AllFeatures --recurse-submodules https://github.com/TaylorJPatterson/LoopWorkspace.git

This places the workspace in ~/Downloads/BuildLoop/Loop-<timestamp>/LoopWorkspace — the same directory structure that L&L's CustomizationSelect script expects.

Step 2: Open in Xcode

BEFORE YOU OPEN Xcode - If you do not want to add compatible L&L Customizations, then proceed to build. Otherwise skip down to the L&L Customization Compatibility section, run that script, then return here to proceed to build.

open ~/Downloads/BuildLoop/$(ls -t ~/Downloads/BuildLoop/ | head -1)/LoopWorkspace/LoopWorkspace.xcworkspace

Step 3: Set your signing team

• Select the Loop target at the top of Xcode
• Go to Signing & Capabilities
• Set your Development Team under the Signing section

Step 4: Build and install

• Connect your iPhone via USB and select it as the build destination
• Press Cmd+R to build and run
• Wait for the build to complete and Loop to install on your device

Step 5: Enable features in the app

Once installed, open Loop on your phone and go to Settings. You'll see three new rows below Therapy Settings:

1. FoodFinder Settings — Tap to configure your AI provider and API key. Once configured, FoodFinder is available from the carb entry screen (the magnifying glass icon).

2. LoopInsights — Tap to enable the feature, configure your AI provider and API key, then open the Dashboard to analyze your therapy settings. LoopInsights will read your glucose, insulin, and carb data and suggest setting adjustments with clinical reasoning.

3. Auto-Apply Presets — Tap to enable and set up rules for automatic preset switching based on motion and activity data from Apple Health.

All three features are OFF by default. Nothing changes about your Loop experience until you choose to enable them individually.

---

How to Build with Browser Build (GitHub Actions)

Step 1: Fork this repository

Click the Fork button at the top of this page to copy this repository to your own GitHub account.

Step 2: Follow the standard Browser Build setup

Follow the Loop Browser Build instructions to set up your certificates and provisioning profiles — the same process as building standard Loop.

Step 3: Run the build

Trigger the GitHub Actions workflow from your fork. All three features are included automatically in the build.

Step 4: Install via TestFlight

Once the build completes, install Loop from TestFlight on your device. Enable the features you want in Settings (same as Step 5 above).

---

Updating to the Latest Version

When new updates are pushed to this branch:

• Xcode builders: Open Terminal, navigate to your LoopWorkspace folder, and run:

  git pull --recurse-submodules

  Then rebuild in Xcode (Cmd+R).

• Browser builders: Sync your fork with this repository and trigger a new build.

---

L&L Customization Compatibility

This branch is compatible with most Loop and Learn customizations. If you already use L&L customizations, you can apply them after cloning this branch, before building in Xcode.

To apply L&L customizations, open Terminal, cd into your cloned LoopWorkspace folder, and run:

/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/loopandlearn/lnl-scripts/main/CustomizationSelect.sh)"

This opens the CustomizationSelect menu where you can pick which L&L customizations to apply. Select the ones you want, then build in Xcode as usual.

Important: Run the customization script after cloning this repo but before building. Our three features are already in the code — the script is only for adding L&L's customizations on top.

Compatibility Table

Patch	Status
2h Lollipop	Compatible
Future Carbs 4h	Compatible
High Carb Limit	Compatible
Low Carb Limit	Compatible
No Auth	Compatible
Override Sensitivity	Compatible
Watch Title	Compatible
LnL Icon	Compatible
Now Line	Compatible
Meal Days (2 day)	Compatible
Meal Week (7 day)	Compatible
Remote Window	Compatible
Profiles	Needs updated patch*
Basal Lock	Needs updated patch*
Negative Insulin	Needs updated patch*

*These three patches conflict with files our features modify (SettingsView.swift and project.pbxproj). Updated patches that work with this branch are being worked on. All other customizations apply cleanly.

---

Requirements

• Xcode 15+ and iOS 17+ (iPhone 15 or newer recommended)
• An AI API key (OpenAI, Anthropic, Google, or any OpenAI-compatible endpoint) for FoodFinder and LoopInsights
• Apple Health permissions for LoopInsights biometric data and AutoPresets motion tracking (optional but recommended)
• Loop 3.10.0 base — this branch is built on the Loop dev branch (v3.10.0)

---

Feature Architecture

All three features are fully isolated with prefixed filenames and subdirectories:

• FoodFinder: FoodFinder_ prefix, FoodFinder/ subdirectories
• LoopInsights: LoopInsights_ prefix, LoopInsights/ subdirectories
• AutoPresets: AutoPresets prefix, AutoPresets/ subdirectory

Only ~38 lines across 3 existing Loop files are modified. Each feature can be enabled or disabled independently at runtime via in-app toggles. Disabling a feature completely removes it from the Settings screen.