iPromise

Product Introduction

1. Definition: iPromise is a macOS-native productivity application (category: Focus & Deep Work Software) that functions as an AI-powered accountability partner. It utilizes real-time screen context analysis and the psychological principle of Body Doubling to enhance user focus during work sessions.
2. Core Value Proposition: iPromise exists to bridge the gap between task planning and task execution. Its core value proposition is providing an intelligent, contextual focus guard that actively intervenes during distraction to help users maintain deep work states and improve productivity metrics.

Main Features

1. AI Focus Buddy with Screen Context Analysis: This is the core technical feature. The application performs local, on-device screen recording (with user permission) to capture the active window context. This visual data is processed locally into screenshots, which are then sent via API to a user-configured AI provider (e.g., OpenAI, Anthropic) for analysis. The AI determines if the on-screen content aligns with the user's declared "promise" or task, classifying the state as "focused" or "distracted." A customizable animated companion (e.g., goose, turtle, comet) provides visual feedback and gentle nudges upon detecting context switches to unapproved applications or websites.
2. Session-Based Focus Protocol with Post-Session Analytics: iPromise operates on timed, single-task sessions. Users initiate a session by declaring a task (via text input, drag-and-drop, or copy-paste) and setting a duration. During the session, a system-wide HUD (Heads-Up Display) shows the focus state. Post-session, the app generates an AI-powered summary, breaking down the session into a 60-segment timeline. It provides an evaluation of focus segments, distraction insights, and suggestions for improvement, creating a quantifiable record of deep work.
3. Privacy-Centric Architecture with iCloud Sync: The application is architected for maximum data privacy. All screen capture and initial processing occur locally on the user's Mac. Screenshots are never stored on iPromise's servers. Users supply their own AI API key, meaning sensitive screen data is transmitted only to their chosen, trusted provider. Non-sensitive session metadata (focus time, summaries, streaks) syncs across the user's Apple devices via iCloud, ensuring a seamless experience without compromising private screen data.

Problems Solved

1. Pain Point: The "planning-execution gap" in personal productivity, where users can list tasks but struggle with sustained, distraction-free execution. It specifically addresses digital distraction and context-switching fatigue during solo deep work sessions.
2. Target Audience: The primary user personas are knowledge workers, remote professionals, students, and creative individuals who perform cognitively demanding tasks on macOS. This includes software developers, writers, researchers, designers, and academics who require prolonged periods of concentrated effort and struggle with self-accountability.
3. Use Cases: Essential for preparing for an important presentation without checking email, writing a report or code without browsing social media, studying for exams while resisting website distractions, or completing any complex, solitary work task that demands uninterrupted cognitive flow.

Unique Advantages

1. Differentiation: Unlike traditional to-do list apps (Todoist, Things) that manage planning, or simple Pomodoro timers (Be Focused) that only manage time, iPromise actively guards focus context. Unlike broad monitoring software, it is a lightweight, user-centric buddy system. It differs from human body-doubling services by being an always-available, AI-driven solution integrated directly into the macOS workspace.
2. Key Innovation: The integration of local screen context analysis with a user-owned AI pipeline for real-time focus state detection. This creates a closed-loop accountability system: the user makes a promise, the AI witnesses the work via screen context, and the buddy provides feedback. The commitment to privacy-by-design, where the app never accesses or stores the analyzed screen data, is a significant technical and ethical innovation in the focus app space.

Frequently Asked Questions (FAQ)

1. How does iPromise's AI focus buddy work without violating my privacy? iPromise uses the macOS screen recording permission to capture context locally on your Mac. Screenshots are processed in memory and sent directly to the AI provider (like OpenAI) using your personal API key. iPromise's servers never receive, see, or store your screen data, ensuring your privacy is maintained.
2. Can I use iPromise with any AI model or provider? Yes, iPromise supports integration with external AI providers via API keys. You can configure it to use services like OpenAI's GPT models, Anthropic's Claude, or other compatible endpoints, giving you control over the AI's capabilities and cost.
3. What happens if I need to switch tasks or applications during a focus session? iPromise is designed for single-task focus. If you switch to an application or website not related to your declared task, the AI will likely classify it as a distraction, and your buddy will nudge you. For legitimate multi-tasking, you should end the current session and start a new one with an updated task promise.
4. Is iPromise available for Windows, iPhone, or iPad? Currently, iPromise is a native application built exclusively for macOS (requires macOS 14.0 or later). It leverages macOS-specific features like the notch HUD and system integrations. The developer, Odd Lab, has not announced plans for Windows or iOS/iPadOS versions at this time.
5. How accurate is the AI at detecting distractions versus focused work? Accuracy depends on your configured AI model and how specifically you define your task. The AI analyzes the visual context of your screen. The app's update notes mention ongoing refinements to "focus state detection to reduce misclassification." For best results, define your task clearly (e.g., "writing in Google Docs" is clearer than "work").