matthiasn/lotti

A private logbook with a staff of personal AI assistants. Agents read what you record and propose what to do next — you approve the changes. End-to-end encrypted sync between your own devices — servers only ever see ciphertext. Local AI optional.

What it solves

Lotti is a distributed system of record for personal memories (voice notes, screenshots, photos, journal entries) that prioritizes user ownership and privacy. It solves the problem of personal data being stored on third-party servers and mined by companies, ensuring that memories remain unaltered, encrypted, and stored exclusively on the user's own devices.

How it works

Lotti uses a local-first architecture where each device holds a full record in a SQLite database. To sync across devices, it uses a zero-trust relay via a Matrix homeserver (Synapse), which only handles encrypted ciphertext.

Key technical mechanisms include:

  • End-to-End Encryption: Uses Megolm and Olm for room messages and AES-256-CTR for attachments, with keys shared only via manual emoji verification.
  • Human-in-the-Loop AI: Personal AI agents read the record to propose tasks, summaries, or updates, but these changes require user confirmation before being applied to the record.
  • Flexible Inference: Users route AI requests to their own configured providers (e.g., zero-data-retention cloud providers or local models like Whisper/Voxtral for speech recognition).
  • Formal Verification: The sync protocol and agent runtime are model-checked using TLA+ to ensure data consistency and prevent loss or duplication during concurrency and crashes.

Who it’s for

It is designed for individuals who want a comprehensive, private system for tracking their life, work, and relationships without relying on a centralized cloud service.

Highlights

  • Distributed Record: No central cloud or account; every paired device is a full peer.
  • Agentic Assistance: Specialized agents for tasks, relationships, goals, projects, and events that propose improvements and summaries.
  • GitHub Integration: Tracks pull requests to link planned work with actual shipments.
  • Privacy-First AI: Supports local inference and zero-data-retention cloud routing.
  • High Reliability: Formally verified sync and recovery logic to prevent data loss.

Related

  • Project
    Open-Curiosity/gini-agent

    A personal AI agent that uses interactive inline controls and live browser handoffs to solve the problem of human-in-the-loop requirements for complex tasks.

  • Project
    thesongzhu/Friday

    A private, local control plane for AI agents that manages routing, verification, and governance across multiple AI vendors to ensure users maintain full control and data ownership.

  • Project
    mikehasa/agentacct

    A local-first agent work ledger that tracks token usage, cost estimates, and activity timelines for AI coding agents using local session logs.

  • Project
    open-software-network/os-clovy

    A private AI workspace for macOS that provides chat, dictation, and meeting notes while keeping data local and routing inference through a verifiable, privacy-preserving API.

  • Project
    codesstar/loci

    Loci is an open‑source, file‑based memory system that lets different AI agents read and write shared context stored locally as Markdown/JSON. It offers personal memory, task scheduling, project indexing, contact logs, a knowledge vault, and a local Node.js dashboard, all without requiring external databases.