AI Frameworks · rowboatlabs

rowboat

Rowboat is an open-source desktop AI assistant that builds a persistent knowledge graph from your work (email, meetings, Slack) and provides built-in surfaces—email client, notes, browser, code editor—to collaborate with AI on tasks. It runs locally, stores data as Markdown, and integrates with external tools via Model Context Protocol.

Talk to DEV.co Discuss Implementation

Source: GitHub — github.com/rowboatlabs/rowboat

15.3k

GitHub stars

1.5k

Forks

TypeScript

Primary language

Apache-2.0

License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

Field	Value
Repository	rowboatlabs/rowboat
Owner	rowboatlabs
Primary language	TypeScript
License	Apache-2.0 — OSI-approved
Stars	15.3k
Forks	1.5k
Open issues	101
Latest release	v0.7.1 (2026-07-07)
Last updated	2026-07-07
Source	https://github.com/rowboatlabs/rowboat

What rowboat is

TypeScript-based Electron desktop app with local-first architecture. Maintains a backlinked knowledge graph in Markdown format, connects to LLMs (Claude, local models via Ollama/LM Studio) via configurable API keys, and extends functionality through MCP server integrations. Supports background agents triggered by events or schedules.

Quickstart

Get the rowboat source

Clone the repository and explore it locally.

terminalbash

1git clone https://github.com/rowboatlabs/rowboat.git2cd rowboat3# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

Knowledge-intensive team workflows

Organizations that need persistent context from email, meetings, and chat. Rowboat's knowledge graph compounds over time, reducing cold-start retrieval overhead and providing auditable, editable work memory.

AI-assisted code development

Teams using Claude Code or similar for parallel coding tasks. Code Mode lets multiple agents work in parallel with full work context, suitable for distributed codebases or exploratory development.

Local-first, privacy-sensitive deployments

Enterprises or individuals requiring on-device processing and storage. All data remains as local Markdown; no cloud lock-in or vendor control. Compatible with self-hosted or local LLMs.

Implementation considerations

Requires manual setup of integrations: Google OAuth, Deepgram API, ElevenLabs, Exa, Composio keys stored locally in JSON config. No centralized credential management.
LLM selection flexibility (Claude, local models, custom providers) means upfront decision on model cost/performance/privacy trade-offs and API key provisioning.
Knowledge graph is editable Markdown in local vault—team processes must establish conventions for graph maintenance, note structure, and version control if shared.
MCP server integrations (Exa, Slack, Linear, etc.) require additional setup per integration; not all workflows may have MCP support available.
Desktop app deployment across Mac/Windows/Linux; monitor Electron version, OS compatibility, and auto-update strategy for team rollout.

When to avoid it — and what to weigh

Real-time multi-user synchronization required — Rowboat is single-user desktop-focused. No built-in collaborative sync, shared workspaces, or conflict resolution for concurrent editing across team members.
Offline-first distributed teams — While local-first, integrations with Gmail, Slack, and external services require internet connectivity. Not designed for intermittent or offline-heavy workflows.
Strict data residency/air-gapped environments — Dependencies on external APIs (Deepgram, ElevenLabs, Exa, Composio) for voice, search, and tools. Local-only mode possible but limited; requires careful API key management.
High-volume production automation at scale — Rowboat targets individual productivity and small team workflows. Not architected for high-throughput batch processing, large-scale automation, or multi-tenant service deployment.

License & commercial use

Apache License 2.0 (Apache-2.0): permissive OSI license. Grants rights to use, modify, distribute, and sublicense the software subject to preservation of copyright notices and license terms. Patent protection included.

Apache-2.0 is OSI-approved and permits commercial use. Source code must remain available and license/copyright notice preserved in distributions. Verify any proprietary integrations or hosted services (rowboatlabs.com, cloud features) for separate commercial terms.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

Signal	Assessment
Maintenance	Active
Documentation	Adequate
License clarity	Clear
Deployment complexity	Moderate
DEV.co fit	Good
Assessment confidence	High

Security considerations

Local-first design reduces cloud exposure; all data stored as plaintext Markdown on user's machine. Threats: API keys stored locally in JSON (no encryption mentioned), browser isolation for assistant access partially mitigates token leakage, meeting audio/transcripts captured locally. No explicit security audit, vulnerability disclosure process, or encryption-at-rest mentioned. Users must manage API key lifecycle and machine security.

Alternatives to consider

Anthropic Claude with persistent chat/file uploads

Hosted, no local infra, simpler setup. Trade-off: cloud-dependent, no explicit knowledge graph, less control over data retention.

OpenAI Canvas / Code Interpreter

Cloud-native collaborative coding and analysis. Missing: persistent knowledge graph, local-first privacy, agent automation on schedules/events.

LangChain + LlamaIndex for custom agents

Developer-friendly framework for building agent workflows and RAG. Trade-off: requires more engineering, no GUI, less packaged integrations or visual collaboration surfaces.

Software development agency

Build on rowboat with DEV.co software developers

Rowboat is production-ready for single-user and small-team AI workflows requiring local data storage and persistent context. Start with the desktop download, test Google/Slack integrations, and pilot background agents or code mode. For enterprise deployment or MCP customization, engage with the Rowboat Labs team or community on Discord.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

rivet hermes-desktop crewAI

Related on DEV.co

Explore the category and the services that help you build with it.

AI Frameworks AI development Custom software Talk to DEV.co

rowboat FAQ

Can I use Rowboat with local models only?

Yes. Rowboat supports local models via Ollama or LM Studio. Bring your own API key or provider; swap anytime. Local deployment means no external LLM API calls if chosen.

Is my data stored in the cloud?

No. All data is stored locally as plain Markdown in your machine's vault. No proprietary formats or hosted lock-in. You can inspect, edit, backup, or delete at any time.

How do I extend Rowboat with custom tools?

Via Model Context Protocol (MCP). Rowboat can connect to MCP servers (Exa, Slack, Linear, GitHub, etc.) or Composio tools. You can build custom MCP servers for internal tools.

Is Rowboat suitable for team use?

Rowboat is single-user desktop-focused. No built-in multi-user sync or conflict resolution. Teams would need per-user installations and manual coordination of shared knowledge if desired.

Work with a software development agency

DEV.co helps companies turn open-source tools like rowboat into production software. Our software development services cover the full lifecycle — architecture, web development, integration, and maintenance — delivered by software developers and web developers who ship. Engage our software development agency to implement or customize it for your ai frameworks stack.

Evaluate Rowboat for Your Team

Talk to DEV.co Schedule a Consultation