Documentation

Welcome to the AI Executive Assistant documentation. This guide will help you set up your own AI-powered meeting workflow, running entirely on your machine.

Recommended Path

While each module works independently, we recommend following this sequence for the best experience:

  1. Foundation — Set up your Obsidian vault and configure your AI provider
  2. Meeting Processing — Enable AI-powered summaries and action item extraction
  3. RAG Search — Add semantic search across your knowledge base
  4. Calendar Integration — Generate proactive meeting briefs
  5. Automation — Set up hands-free processing with n8n

Quick Start

Want to skip the details? Use our Setup Wizard to generate a customized configuration based on your preferences.

Prerequisites

Before you begin, you'll need:

  • Obsidian — Free markdown-based note-taking app (download)
  • Python 3.11+ — For running processing scripts
  • Docker — For Qdrant vector database and n8n (optional modules)
  • An AI Provider — Ollama (local), OpenAI, Claude, or Vertex AI

Privacy

This system is designed with privacy as a core principle. By default:

  • All processing happens on your machine
  • Your notes never leave your computer unless you choose a cloud AI provider
  • You can use Ollama for fully local AI processing
  • No telemetry, no tracking, no accounts required