rowboatlabs/rowboatPublic

Open-source AI coworker, with memory

AI summary: A desktop AI coworker that indexes your local environment to provide deeply contextual, proactive automation.

Star on GitHub
Stars
17K
+19 today
Forks
1.7K
Watchers
90
Open issues
71
Open PRs
63
Contributors
~16
Commits
2.2K
Branches
268

TypeScriptApache-2.0Created Jan 13, 2025Last push 1d agoLatest release v0.8.3+111 stars this week+111 this month

Star history

since Jan 26, 2025
05K10K15KJan 2025Jul 2025Jan 2026Aug 2026
17K stars as of Aug 6, 2026, tracked back to Jan 26, 2025. Historical curve reconstructed from public GitHub event archives, calibrated to the current total.

Signals and awards

derived from tracked data

  • Widely adopted

    16,988 stars

  • Actively maintained

    Pushed within 48 hours

  • Permissive license

    Apache-2.0

  • Continuous integration

    Automated checks passing

  • Repeat trending

    3 trending appearances

What rowboat does

Rowboat is a desktop-native AI assistant designed to solve the problem of fragmented context in modern knowledge work. Unlike standard chat interfaces, it actively indexes your local work environment—your files, applications, and history—to maintain an ongoing memory of what you are doing. It provides built-in surfaces to act on this context, effectively functioning as a multi-agent orchestration layer right on your machine. The system integrates tightly with your operating system, allowing it to automate tasks, recall past actions, and assist in coding or writing with a deep understanding of your personal workspace. By running locally, it ensures the AI has the necessary granular context that cloud-only tools inherently lack.

Knowledge workers, developers, and professionals who want a deeply integrated AI assistant that understands their local files and specific workflows. Requires installation on a local desktop environment.

  • Local context indexing: Continuously scans and indexes your local files and workflows to build a persistent memory of your work.
  • Desktop native integration: Operates directly on your OS, allowing it to interact with local applications and file systems.
  • Multi-agent architecture: Orchestrates various AI models beneath the surface to handle different types of automation tasks.
  • Actionable built-in surfaces: Provides specific UI elements and triggers allowing the AI to execute commands, not just chat.
  • Persistent work memory: Recalls previous projects, code snippets, and decisions to inform current tasks automatically.

Where teams use it

Context-aware coding assistance

Software developers can ask Rowboat to refactor code, knowing the AI already has their entire local repository indexed and understood.

Automated workflow execution

Knowledge workers can instruct the AI to perform complex, multi-step actions across different local applications based on historical patterns.

Intelligent local search

Users can query their entire workspace history to instantly surface relevant documents, conversations, or code they worked on weeks ago.

Proactive project management

The AI can suggest next steps or identify missing context for a project by analyzing the current state of the local file system.

Getting started: Visit https://www.rowboatlabs.com/

README

main branch
rowboat-github-2

Rowboat

A desktop AI coworker with a memory of your work and built-in surfaces to act on it.

rowboatlabs/rowboat | Trendshift

Website Discord Twitter Y Combinator

Rowboat indexes your work into a living knowledge graph and uses that to get work done on your machine. It includes work surfaces for collaborating with AI: email client, notes, browser, code mode, meeting note taker, and workspaces for different projects.

Download latest for Mac/Windows/Linux: Download

Rowboat Apps to Code demo

Demo - apps to code · Demo - knowledge graph

⭐ If you find Rowboat useful, please star the repo. It helps more people find it.

Overview

Brain

Rowboat indexes email, meetings, slack and assistant conversations into a living Obsidian-style backlinked knowledge graph. 		Brain graph screenshot

Email

The built-in email client sorts emails into important and everything else. Rowboat automatically drafts responses for important email using all the work context. 		Email screenshot

Background agents

You can set up background agents that run on events like new email or on schedule like every day at 8am. They can connect to tools, search the web, use the browser and write code using Claude Code or Codex. 		Background agents screenshot

Built-in Browser

Rowboat includes a browser that lets you and assistant collaborate on web tasks. Because it's isolated from your main browser, you can log in only to the accounts that want the assistant to access. 		Browser screenshot

Meeting Notes

A local meeting note-taker that taps into mic & speaker, produces live transcript and summarizes the meeting in a markdown file and updates the knowledge graph. 		Meeting notes screenshot

Code Mode

Code mode lets you spin up parallel coding agents with Claude Code or Codex, and have Rowboat drive them with all the work context where needed. 		Code mode screenshot

Apps

You can build your own work surfaces inside Rowboat — they get access to all the tools and integrations, and you can share them with other people. 		Apps screenshot

Integrations

Includes one-click integrations to most popular products. 		Integrations screenshot

Installation

Download latest for Mac/Windows/Linux: Download

All release files: https://github.com/rowboatlabs/rowboat/releases/latest

Google setup

To connect Google services (Gmail, Calendar, and Drive), follow Google setup.

Voice input

To enable voice input and voice notes (optional), add a Deepgram API key in ~/.rowboat/config/deepgram.json

Voice output

To enable voice output (optional), add an ElevenLabs API key in ~/.rowboat/config/elevenlabs.json

Web search

To use Exa research search (optional), add the Exa API key in ~/.rowboat/config/exa-search.json

External tools

To enable external tools (optional), you can add any MCP server or use Composio tools by adding an API key in ~/.rowboat/config/composio.json

All API key files use the same format:

{
  "apiKey": "<key>"
}

How it’s different

Most AI tools reconstruct context on demand by searching transcripts or documents.

Rowboat maintains long-lived knowledge instead:

  • context accumulates over time
  • relationships are explicit and inspectable
  • notes are editable by you, not hidden inside a model
  • everything lives on your machine as plain Markdown

The result is memory that compounds, rather than retrieval that starts cold every time.

Bring your own model

Rowboat works with the model setup you prefer:

  • Local models via Ollama or LM Studio
  • Hosted models (bring your own API key/provider)
  • Swap models anytime — your data stays in your local Markdown vault

Extend Rowboat with tools (MCP)

Rowboat can connect to external tools and services via Model Context Protocol (MCP). That means you can plug in (for example) search, databases, CRMs, support tools, and automations - or your own internal tools.

Examples: Exa (web search), Twitter/X, ElevenLabs (voice), Slack, Linear/Jira, GitHub, and more.

Local-first by design

  • All data is stored locally as plain Markdown
  • No proprietary formats or hosted lock-in
  • You can inspect, edit, back up, or delete everything at any time
View on GitHub

Recent activity

commits and pull requests

Recent open issues

view all

Releases and announcements

158 total
  1. v0.8.3v0.8.3Jul 29, 2026818 downloads

    ## What's new **✨ Ask Rowboat from anywhere.** Press ⌥⇧Space in any app and a Spotlight-style quick-ask bar appears. Type your question — or hold right ⌘ to speak it — and the answer streams right back into the bar, with a live status line showing what the agent is doing. The question lands in your current Rowboat chat, so nothing is lost: hit "Open in Rowboat" to jump into the full conversation, or "+" to start a fresh one. You can also toggle voice responses and screen sharing without leaving the bar. On macOS Tahoe the bar is rendered in native Liquid Glass. **🎙️ Push-to-talk calls.** Hold right ⌘ to talk during a call — release to send. Calls gained a popout text input, clearer controls with tooltips, and the popout and quick-ask bar now float above fullscreen apps instead of getting lost behind them. **🔐 Smoother permission prompts.** Screen Recording permission is settled once at call start for every mode, permission alerts are audible and bring themselves into focus, and screen share in the packaged app is now reliable. ### Also in this release - Calls no longer lose responses — dropped utterances and voiceless replies are fixed, and every reply is spoken fr

  2. v0.8.2v0.8.2Jul 28, 2026pre-release32 downloads

    ## What's new **✨ Ask Rowboat from anywhere.** Press ⌥⇧Space in any app and a Spotlight-style quick-ask bar appears. Type your question — or hold right ⌘ to speak it — and the answer streams right back into the bar, with a live status line showing what the agent is doing. The question lands in your current Rowboat chat, so nothing is lost: hit "Open in Rowboat" to jump into the full conversation, or "+" to start a fresh one. You can also toggle voice responses and screen sharing without leaving the bar. On macOS Tahoe the bar is rendered in native Liquid Glass. **🎙️ Push-to-talk calls.** Hold right ⌘ to talk during a call — release to send. Calls gained a popout text input, clearer controls with tooltips, and the popout and quick-ask bar now float above fullscreen apps instead of getting lost behind them. **🔐 Smoother permission prompts.** Screen Recording permission is settled once at call start for every mode, permission alerts are audible and bring themselves into focus, and screen share in the packaged app is now reliable. ### Also in this release - Calls no longer lose responses — dropped utterances and voiceless replies are fixed, and every reply is spoken fr

  3. v0.8.1v0.8.1Jul 27, 2026pre-release18 downloads
  4. v0.8.0v0.8.0Jul 24, 20263.7K downloads

    ## What's new **🎛️ One place for models.** Provider setup and model choice now live in a single redesigned settings experience: connect a provider, browse every available model in a split-view picker, and get per-task recommendations applied automatically. Background tasks (notes, titles, live notes) can run on their own models or simply inherit your assistant's. **📊 Charts, right in the chat.** Ask for a comparison or a trend and the assistant can now draw inline charts directly in the conversation — no export, no external tool. **✍️ Chats name themselves.** New conversations get a short, meaningful title from your first message, so your history reads like a table of contents instead of a wall of "New chat". **🪟 Signed Windows builds.** Windows releases are now code-signed via Azure Trusted Signing — fewer scary install warnings, and release notes now show on the Windows update card too. **🏃 Longer runs, graceful finishes.** The assistant can take up to 50 model calls per turn by default (configurable up to 500, per workflow) — and when it does hit the limit, it wraps up with a real answer instead of stopping mid-thought. ### Also in this release - Billing

  5. v0.7.9v0.7.9Jul 23, 2026pre-release33 downloads

When work happens

weekday and hour
SunMonTueWedThuFriSat036912151821Sun 0:00 — 4 commitsSun 1:00 — 1 commitsSun 2:00 — 2 commitsSun 3:00 — 1 commitsSun 4:00 — 0 commitsSun 5:00 — 0 commitsSun 6:00 — 2 commitsSun 7:00 — 3 commitsSun 8:00 — 2 commitsSun 9:00 — 0 commitsSun 10:00 — 2 commitsSun 11:00 — 6 commitsSun 12:00 — 4 commitsSun 13:00 — 3 commitsSun 14:00 — 4 commitsSun 15:00 — 6 commitsSun 16:00 — 7 commitsSun 17:00 — 0 commitsSun 18:00 — 3 commitsSun 19:00 — 3 commitsSun 20:00 — 1 commitsSun 21:00 — 2 commitsSun 22:00 — 1 commitsSun 23:00 — 7 commitsMon 0:00 — 2 commitsMon 1:00 — 2 commitsMon 2:00 — 1 commitsMon 3:00 — 0 commitsMon 4:00 — 1 commitsMon 5:00 — 0 commitsMon 6:00 — 4 commitsMon 7:00 — 2 commitsMon 8:00 — 3 commitsMon 9:00 — 7 commitsMon 10:00 — 15 commitsMon 11:00 — 18 commitsMon 12:00 — 17 commitsMon 13:00 — 13 commitsMon 14:00 — 31 commitsMon 15:00 — 29 commitsMon 16:00 — 30 commitsMon 17:00 — 20 commitsMon 18:00 — 6 commitsMon 19:00 — 7 commitsMon 20:00 — 16 commitsMon 21:00 — 27 commitsMon 22:00 — 21 commitsMon 23:00 — 18 commitsTue 0:00 — 15 commitsTue 1:00 — 14 commitsTue 2:00 — 7 commitsTue 3:00 — 4 commitsTue 4:00 — 0 commitsTue 5:00 — 0 commitsTue 6:00 — 5 commitsTue 7:00 — 4 commitsTue 8:00 — 7 commitsTue 9:00 — 7 commitsTue 10:00 — 10 commitsTue 11:00 — 23 commitsTue 12:00 — 28 commitsTue 13:00 — 30 commitsTue 14:00 — 41 commitsTue 15:00 — 42 commitsTue 16:00 — 38 commitsTue 17:00 — 35 commitsTue 18:00 — 15 commitsTue 19:00 — 20 commitsTue 20:00 — 31 commitsTue 21:00 — 18 commitsTue 22:00 — 23 commitsTue 23:00 — 23 commitsWed 0:00 — 17 commitsWed 1:00 — 12 commitsWed 2:00 — 2 commitsWed 3:00 — 1 commitsWed 4:00 — 0 commitsWed 5:00 — 1 commitsWed 6:00 — 3 commitsWed 7:00 — 6 commitsWed 8:00 — 2 commitsWed 9:00 — 6 commitsWed 10:00 — 17 commitsWed 11:00 — 16 commitsWed 12:00 — 14 commitsWed 13:00 — 21 commitsWed 14:00 — 23 commitsWed 15:00 — 20 commitsWed 16:00 — 24 commitsWed 17:00 — 12 commitsWed 18:00 — 16 commitsWed 19:00 — 11 commitsWed 20:00 — 20 commitsWed 21:00 — 16 commitsWed 22:00 — 19 commitsWed 23:00 — 24 commitsThu 0:00 — 12 commitsThu 1:00 — 11 commitsThu 2:00 — 7 commitsThu 3:00 — 0 commitsThu 4:00 — 0 commitsThu 5:00 — 1 commitsThu 6:00 — 4 commitsThu 7:00 — 7 commitsThu 8:00 — 3 commitsThu 9:00 — 10 commitsThu 10:00 — 12 commitsThu 11:00 — 26 commitsThu 12:00 — 13 commitsThu 13:00 — 18 commitsThu 14:00 — 17 commitsThu 15:00 — 21 commitsThu 16:00 — 23 commitsThu 17:00 — 23 commitsThu 18:00 — 14 commitsThu 19:00 — 15 commitsThu 20:00 — 11 commitsThu 21:00 — 11 commitsThu 22:00 — 19 commitsThu 23:00 — 19 commitsFri 0:00 — 19 commitsFri 1:00 — 17 commitsFri 2:00 — 14 commitsFri 3:00 — 5 commitsFri 4:00 — 0 commitsFri 5:00 — 3 commitsFri 6:00 — 2 commitsFri 7:00 — 4 commitsFri 8:00 — 7 commitsFri 9:00 — 11 commitsFri 10:00 — 22 commitsFri 11:00 — 21 commitsFri 12:00 — 16 commitsFri 13:00 — 22 commitsFri 14:00 — 19 commitsFri 15:00 — 21 commitsFri 16:00 — 31 commitsFri 17:00 — 27 commitsFri 18:00 — 10 commitsFri 19:00 — 18 commitsFri 20:00 — 9 commitsFri 21:00 — 9 commitsFri 22:00 — 12 commitsFri 23:00 — 21 commitsSat 0:00 — 20 commitsSat 1:00 — 17 commitsSat 2:00 — 3 commitsSat 3:00 — 1 commitsSat 4:00 — 0 commitsSat 5:00 — 0 commitsSat 6:00 — 0 commitsSat 7:00 — 1 commitsSat 8:00 — 3 commitsSat 9:00 — 8 commitsSat 10:00 — 12 commitsSat 11:00 — 2 commitsSat 12:00 — 15 commitsSat 13:00 — 6 commitsSat 14:00 — 7 commitsSat 15:00 — 2 commitsSat 16:00 — 2 commitsSat 17:00 — 3 commitsSat 18:00 — 5 commitsSat 19:00 — 2 commitsSat 20:00 — 2 commitsSat 21:00 — 1 commitsSat 22:00 — 3 commitsSat 23:00 — 4 commits
Commit volume by weekday and hour (UTC). Larger dots mean more commits.

Trending appearances

DateListRankStars gained
Feb 16, 2026daily#15+173
Feb 15, 2026daily#7+316
Feb 14, 2026daily#10+204

Related repositories

  • public-apis/public-apis

    A collective list of free APIs

    454.9K stars · Python

  • freeCodeCamp/freeCodeCamp

    freeCodeCamp.org's open-source codebase and curriculum. Learn math, programming, and computer science for free.

    453.6K stars · TypeScript

  • openclaw/openclaw

    Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

    385.5K stars · TypeScript

  • openclaw/openclaw

    Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

    384.4K stars · TypeScript

  • openclaw/openclaw

    Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

    384.4K stars · TypeScript

  • openclaw/openclaw

    Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

    384.4K stars · TypeScript