rowboatlabs/rowboat

Open-source AI coworker, with memory

What it solves

Rowboat addresses the problem of "cold start" AI retrieval, where tools must reconstruct context from scratch for every request. It provides a desktop AI coworker that maintains long-lived, compounding memory of a user's work, ensuring that context accumulates over time rather than being retrieved on demand.

How it works

Rowboat indexes data from emails, meetings, Slack, and assistant conversations into a local, backlinked knowledge graph stored as plain Markdown files. It integrates various "work surfaces"—including a built-in email client, browser, and meeting note-taker—to act on this data. Users can deploy background agents to handle scheduled tasks or event-driven triggers, and connect external tools via the Model Context Protocol (MCP). It supports both local models (via Ollama or LM Studio) and hosted models via API keys.

Who it’s for

It is designed for professionals who want an AI assistant with deep, persistent context of their professional life and workflows, and who prefer a local-first approach to data storage and privacy.

Highlights

  • Living Knowledge Graph: Automatically indexes work communications into an Obsidian-style backlinked graph.
  • Local-First Storage: All data is stored on the user's machine as plain Markdown, avoiding proprietary lock-in.
  • Integrated Work Surfaces: Includes a dedicated email client, isolated browser, and a local meeting note-taker with live transcription.
  • Agentic Capabilities: Supports background agents for automation and "Code Mode" for parallel coding agents using Claude Code or Codex.
  • Extensible: Connects to external services and databases through the Model Context Protocol (MCP).

Related

  • Dispatch
    Rowboat Open-Source Local-First AI Coworker – Features, Architecture, and Community Feedback

    Rowboat is an open‑source desktop AI coworker that stores all data locally as Markdown, builds a persistent knowledge graph, and provides built‑in work surfaces like email, notes, browser, and code agents.

  • Project
    siddsachar/row-bot

    A local-first desktop AI assistant that uses agent orchestration and durable memory to perform complex work across local files, repositories, and external tools.

  • Project
    eggent-ai/eggent

    A local-first AI workspace and orchestration layer that provides project-scoped agents with persistent memory, files, and pipeline capabilities.

  • Project
    cabinetai/cabinet

    A self-hosted, local-first knowledge base and AI agent orchestration platform that stores all data as Markdown files on disk to provide persistent memory for AI teams.

  • Project
    shiwenwen/hope-agent

    A local-first personal AI agent that autonomously executes long-term goals, manages a persistent knowledge base, and controls desktop/browser interfaces.