Open-source macOS meeting-notes app that records mic and system audio without a bot, transcribes locally or via cloud APIs, and writes summaries from your notes plus the transcript.
- Stars
- 1,593
- Forks
- 182
- Open issues
- 89
- Last commit
- 20 hours ago
What is it?
Minutes records and transcribes meetings, calls, voice memos, and dictation on your own device, then stores each conversation as a Markdown file with YAML frontmatter in a local folder such as ~/meetings/. Capture, transcription, and storage all run locally; text only leaves the machine when you send meeting context to a cloud AI assistant or summarizer.
The project is built for AI agents. An MCP server exposes tools and resources so Claude Code, Codex, Cursor, Gemini CLI, Claude Desktop, and other MCP clients can search and read your meeting history. It also ships a CLI with commands for recording, processing, search, and import, plus a TypeScript SDK for reading meeting files without MCP.
The desktop app adds menu bar capture, dictation with personal spellings and snippets, and features such as Recall and Coach. Meetings are plain Markdown, so they open directly in Obsidian or any text tool. Sensitive meetings can be saved without audio, and consent reminders help with disclosure.
Key features
- 01Transcribes meetings, calls, and voice memos on-device.
- 02Saves each conversation as Markdown with structured YAML frontmatter for action items and decisions.
- 03Exposes an MCP server so Claude Code, Codex, Cursor, and other clients can search meetings.
- 04Provides a CLI with commands for recording, processing, search, and import.
- 05Includes a desktop app with menu bar capture and dictation.
- 06Offers a TypeScript SDK for reading meeting files directly.
Use cases
- Asking an AI agent what was decided across several past meetings.
- Keeping meeting notes in a local Obsidian vault or personal knowledge base.
- Recording calls and voice memos without sending audio to a hosted service.
- Migrating meeting archives from hosted note tools into local Markdown files.
Quick start
- 1
Install the standalone CLI with Homebrew
brew install silverstein/tap/minutes - 2
Add the MCP server to Claude Code
claude mcp add minutes -- npx -y minutes-mcp - 3
Seed five sample meetings to try it without a microphone
minutes demo --full
Things to know
- Some features, such as automatic Teams and Zoom call recording and dictation cursor context, are macOS only.
- Local-first capture does not apply to a cloud AI assistant or summarizer you choose, which receives the context you authorize.
In collections
Topics
FAQ
- What is minutes?
- Open-source, local-first meeting and voice memo recorder that transcribes on-device into Markdown files that AI agents like Claude Code can search via MCP. Minutes records and transcribes meetings, calls, voice memos, and dictation on your own device, then stores each conversation as a Markdown file with YAML frontmatter in a local folder such as ~/meetings/. Capture, transcription, and storage all run locally; text only leaves the machine when you send meeting context to a cloud AI assistant or summarizer.
- Is minutes free to use? What license does it have?
- minutes is open source under the MIT license. Check the license terms before using it in commercial products.
- Is minutes still maintained?
- Actively maintained. The last commit to minutes was on Nov 6, 2026.
- What can I use minutes for?
- Asking an AI agent what was decided across several past meetings. Keeping meeting notes in a local Obsidian vault or personal knowledge base. Recording calls and voice memos without sending audio to a hosted service. Migrating meeting archives from hosted note tools into local Markdown files.
- How do I install minutes?
- Install the standalone CLI with Homebrew: brew install silverstein/tap/minutes Add the MCP server to Claude Code: claude mcp add minutes -- npx -y minutes-mcp Seed five sample meetings to try it without a microphone: minutes demo --full
- Where can I get minutes?
- The source code and installation instructions are on GitHub: https://github.com/silverstein/minutes
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Open-source desktop dictation app that turns speech into text at your cursor, with local Whisper/Parakeet models or cloud providers.
Privacy-first AI meeting assistant that records, transcribes and summarizes meetings locally on macOS and Windows, with optional Ollama or cloud LLM summaries.
Open-source desktop AI note taker that transcribes meetings and voice notes without a bot joining the call, using local or cloud AI.
A macOS meeting note-taker that transcribes both sides of a call locally and surfaces talking points from your own notes.
Self-hosted, offline audio and video transcription app with speaker detection, chat with transcripts, and a folder watcher for automation.