<2-3 sentences summarising what the project addresses and solves, the target audience, and the benefit.> Lotti is a private, local-first AI-powered context manager designed for individuals seeking to leverage AI for productivity without compromising data privacy. It enables users to manage tasks, notes, and audio recordings, offering intelligent summaries and chat capabilities, with all data stored securely on local devices.

How It Works

Lotti employs a local-first architecture, ensuring complete data ownership. Users can configure various AI providers, including cloud services like OpenAI and Anthropic, or run models entirely offline via Ollama. This approach prioritizes privacy by allowing selective data sharing only for specific inference calls, managed through user-provided API keys. The system utilizes Flutter for a cross-platform user interface, local Whisper for audio transcription, and Matrix for end-to-end encrypted synchronization between devices.

Quick Start & Requirements

Installation involves cloning the repository and running make deps after setting up Flutter (FVM is recommended). Beta testing is available via TestFlight for iOS/macOS, with Linux builds planned for Flatpak. Detailed setup, development workflows, and documentation are available in the repository. Running local AI models requires capable hardware.

Highlighted Details

• AI-Powered Intelligence: Features smart summaries, audio transcription (local Whisper or cloud), context recaps for tasks, intelligent checklists, and conversational data querying.
• Comprehensive Tracking: Supports full task lifecycle management, audio recording, time tracking, journaling, habit monitoring, health data import (e.g., Apple Health), and custom metrics.
• Privacy & Control: Employs local-only storage, end-to-end encrypted sync via Matrix, selective AI usage configuration, and GDPR-compliant European AI provider options.
• Flexible AI Integration: Seamlessly integrates with OpenAI, Anthropic Claude, Google Gemini, Ollama, and any OpenAI-compatible API endpoints.

Maintenance & Community

Development is conducted publicly on GitHub, with updates and project background shared via Substack. The project encourages contributions and outlines standards for participation.

Licensing & Compatibility

Lotti is distributed under an open-source license, with the specific type detailed in the repository's LICENSE file. Its Flutter-based architecture ensures compatibility across iOS, macOS, Android, Linux, and Windows.

Limitations & Caveats

The project is currently in beta, with TestFlight builds available for select users. Encrypted synchronization necessitates a Matrix account. Achieving full offline AI capabilities depends on having sufficiently powerful hardware. The precise open-source license terms are not explicitly detailed in the provided README text.