ef8584c05f
The renderer previously filtered run events by the active runId, which meant run-processing-start for a newly created or non-active run could be dropped. That caused the thinking indicator to miss on subsequent chats and lost live streaming context when switching between runs. This change keeps only minimal background state: - a Set of runIds currently processing (driven by run-processing-start/end) - a per-run streaming buffer for assistant/reasoning deltas All run events are now observed; non-active runs only update the minimal structures. When a run is selected, its conversation is loaded as before and any in-progress buffer is used to seed the streaming UI. Buffers are cleared on final assistant message, run end, stop, or error. The chat history list now shows a processing indicator per run. Tests: - cd apps/x && npm run deps - cd apps/x/apps/renderer && npm run build |
||
|---|---|---|
| .github/workflows | ||
| apps | ||
| assets | ||
| .env.example | ||
| .gitattributes | ||
| .gitignore | ||
| build-electron.sh | ||
| CLAUDE.md | ||
| docker-compose.yml | ||
| Dockerfile.qdrant | ||
| google-setup.md | ||
| LICENSE | ||
| README.md | ||
| start.sh | ||
Rowboat
An open-source, local-first AI coworker with memory for everyday work
Rowboat connects to your email and meeting notes, builds a long-lived knowledge graph, and uses that context to help you get work done - privately, on your machine.
You can do things like:
Build me a deck about our next quarter roadmap→ generates a PDF using context from your knowledge graphPrep me for my meeting with Alex→ pulls past decisions, open questions, and relevant threads into a crisp brief (or a voice note)- Visualize, edit, and update your knowledge graph anytime (it’s just Markdown)
- Record voice memos that automatically capture and update key takeaways in the graph
Download latest for Mac/Windows/Linux: Download
Demo
Installation
Download latest for Mac/Windows/Linux: Download
All release files: https://github.com/rowboatlabs/rowboat/releases/latest
What it does
Rowboat is a local-first AI coworker that can:
- Remember the important context you don’t want to re-explain (people, projects, decisions, commitments)
- Understand what’s relevant right now (before a meeting, while replying to an email, when writing a doc)
- Help you act by drafting, summarizing, planning, and producing real artifacts (briefs, emails, docs, PDF slides)
Under the hood, Rowboat maintains an Obsidian-compatible vault of plain Markdown notes with backlinks — a transparent “working memory” you can inspect and edit.
Integrations
Rowboat builds memory from the work you already do, including:
- Gmail (email)
- Granola (meeting notes)
- Fireflies (meeting notes)
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.
What you can do with it
- Meeting prep from prior decisions, threads, and open questions
- Email drafting grounded in history and commitments
- Docs & decks generated from your ongoing context (including PDF slides)
- Follow-ups: capture decisions, action items, and owners so nothing gets dropped
- On-your-machine help: create files, summarize into notes, and run workflows using local tools (with explicit, reviewable actions)
Background agents
Rowboat can spin up background agents to do repeatable work automatically - so routine tasks happen without you having to ask every time.
Examples:
- Draft email replies in the background (grounded in your past context and commitments)
- Generate a daily voice note each morning (agenda, priorities, upcoming meetings)
- Create recurring project updates from the latest emails/notes
- Keep your knowledge graph up to date as new information comes in
You control what runs, when it runs, and what gets written back into your local Markdown vault.
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
Looking for Rowboat Web Studio?
If you’re looking for Rowboat web Studio, start here.