Meeting Notes & Action Intelligence

A portable Windows desktop application for capturing, structuring, summarizing, and operationalizing meeting notes. Built with Tauri 2 + Rust + React + TypeScript.

Features

• Raw Note Capture — Type notes live, paste transcripts (Teams/Zoom/Meet), import text/markdown files
• AI-Powered Processing — Clean grammar, generate structured minutes, extract action items, decisions, deliverables, risks, open questions
• Human-in-the-Loop — Review and accept/reject/edit AI suggestions before committing
• Action Item Tracking — Full CRUD with owner, due date, priority, status, confidence scoring, tentative flags
• Full-Text Search — FTS5-powered search across all meetings and notes
• Export — Markdown, CSV, JSON export with save-to-file dialog
• Portable Mode — Runs as a portable app without installation
• Secure — API keys stored in Windows Credential Store, never in config files
• Offline-First — Core features work without internet; AI features degrade gracefully

Tech Stack

Layer	Technology
Desktop Shell	Tauri 2
Backend	Rust (rusqlite, keyring, reqwest)
Frontend	React 19, TypeScript, Vite
Styling	Tailwind CSS 4
Database	SQLite with FTS5
AI	OpenAI-compatible API (Ollama, LM Studio, OpenAI)

Prerequisites

• Rust 1.70+ with cargo
• Node.js 18+ with npm
• Windows 10 22H2 or later (for WebView2)

Setup

# Clone the repo
git clone <repo-url> meeting-notes
cd meeting-notes

# Install frontend dependencies
npm install

# Run in development mode
npm run tauri dev

The first cargo build will take a few minutes to download and compile Rust dependencies.

Build for Production

# Build optimized release
npm run tauri build

The installer/executable is output to src-tauri/target/release/bundle/.

Portable Mode

The app supports portable distribution — no installer required.

How it works

At startup, the app checks for these markers next to the executable:

1. A file named portable (can be empty)
2. OR a folder named data/

If either exists, the app runs in portable mode and stores all data inside <exe_dir>/data/:

meeting-notes.exe
portable              ← marker file (or just create a data/ folder)
data/
  db/                 ← SQLite database
  config/             ← settings.json
  exports/            ← exported files
  logs/               ← application logs

If neither marker exists, data is stored in %APPDATA%/meeting-notes/.

Creating a portable distribution

1. Build the release: npm run tauri build
2. Copy the .exe from src-tauri/target/release/
3. Create a portable file (or data/ folder) next to it
4. Distribute the folder

AI Provider Configuration

Mock Provider (Default)

No configuration needed. Returns realistic sample data for development and testing.

Ollama (Local)

1. Install Ollama and pull a model (e.g., ollama pull llama3)
2. In Settings, set:
   • Provider: openai-compatible
   • Endpoint: http://localhost:11434/v1
   • Model: llama3

LM Studio (Local)

1. Start LM Studio server
2. In Settings, set:
   • Provider: openai-compatible
   • Endpoint: http://localhost:1234/v1
   • Model: (your loaded model name)

OpenAI API

1. In Settings, set:
   • Provider: openai-compatible
   • Endpoint: https://api.openai.com/v1
   • Model: gpt-4o-mini (or your preferred model)
2. Enter your API key in the API Key section (stored in Windows Credential Store)

Project Structure

meeting-notes/
├── src/                        # React frontend
│   ├── components/
│   │   ├── ui/                 # Base UI components (Button, Badge, Card, Tabs)
│   │   ├── layout/             # AppLayout, Sidebar
│   │   ├── meetings/           # NoteEditor
│   │   ├── actions/            # ActionItemRow
│   │   ├── ai/                 # ExtractionReview
│   │   └── search/
│   ├── pages/                  # Dashboard, Meetings, MeetingDetail, ActionItems, Settings
│   ├── hooks/                  # useAutosave
│   ├── lib/                    # api.ts (Tauri invoke wrapper), utils.ts
│   ├── types/                  # TypeScript interfaces
│   └── styles/                 # Tailwind globals
├── src-tauri/                  # Rust backend
│   └── src/
│       ├── app/commands/       # Tauri command handlers (thin layer)
│       ├── domain/             # Domain models (Meeting, ActionItem, Minutes, etc.)
│       ├── services/           # Business logic
│       ├── db/                 # SQLite repos, migrations, FTS5 search
│       ├── ai/                 # Provider trait, MockProvider, OpenAiCompatibleProvider
│       ├── config/             # Portable mode, app config
│       ├── export/             # Markdown, CSV, JSON export
│       └── utils/              # Error types
└── ARCHITECTURE.md             # Full architecture documentation

Database

SQLite with WAL mode for performance. Schema includes:

• meetings — Meeting sessions with status tracking
• raw_notes — Original unprocessed notes (source of truth)
• processed_minutes — AI-generated structured minutes
• action_items — Tracked commitments with confidence scores
• decisions, deliverables, risks_blockers, open_questions — Extracted entities
• ai_processing_runs — Audit trail for AI operations
• meetings_fts — FTS5 full-text search index

Migrations run automatically on startup.

Security & Privacy

• All data stored locally by default
• API keys stored in Windows Credential Store via keyring crate
• Keys never appear in logs, config files, or exports
• When using API-based AI providers, note content is sent to the configured endpoint
• No telemetry or analytics

License

MIT