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 private, local-first logbook designed to track work and personal life without sacrificing privacy. It solves the problem of "honest" time tracking by separating the intended plan (tasks) from the actual reality (time records), preventing the pressure to pretend a day went according to plan.

How it works

Lotti stores data in a local SQLite database on the user's device. It employs a "human-in-the-loop" architecture where AI agents operate in a separate "agentic database" to prevent them from directly modifying the system of record. Agents can read user data and propose changes (like summaries or next steps), but these changes must be explicitly approved by the user before they are written to the user database.

Syncing is handled via end-to-end encrypted Matrix/Synapse homeservers, ensuring that the relay server only ever holds ciphertext. For AI, Lotti allows users to route specific categories of their life to different providers (local models, frontier models, or EU-based open-weight models) based on their privacy needs.

Who it’s for

It is for individuals who want a high-degree of sovereignty over their personal data, people who prefer local-first software, and those who need a detailed, honest record of their work, habits, and health data.

Highlights

  • Separation of Intent and Reality: Keeps tasks and actual time records as separate facts.
  • Human-in-the-Loop AI: Agents propose changes; users decide, enforced by separate on-disk databases.
  • Sovereign Data: Local SQLite storage with no telemetry and end-to-end encrypted sync.
  • Flexible AI Routing: Route different life categories to different LLM providers (e.g., local for private, cloud for work).
  • Local Transcription: Offline audio recording and transcription using Whisper or Voxtral.
  • Impact Tracking: Reports token usage, cost, energy, and CO2e for providers that disclose this data.

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
    FutureUniant/WorkShadow

    A local-first desktop work log application that uses AI for semantic search, automated report generation, and Q&A over personal work history while keeping data on the user's machine.

  • Project
    fastrepl/anarlog

    A privacy-first, local-first AI meeting notepad that records audio directly from your device to provide transcriptions and summaries without needing a bot to join the call.

  • 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
    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.