Webinar Crewai Qdrant Obsidian — Development skill for Claude Code

webinar-email-assistant-crewai-obsidian

This repository contains materials from the hands-on webinar "[Building Intelligent Agentic RAG with CrewAI and Qdrant](https://www.youtube.com/watch?v=soGB3UowTZ0)". It implements an agentic RAG system that uses your existing [Obsidian](https://obsidian.md/) Vault as a knowledge base to draft emails in Gmail Inbox.

Software Stack

The project is built using Python and integrates with external services through their APIs. **The agentic behaviors are implemented using [CrewAI](https://www.crewai.com/), with [Qdrant](https://qdrant.tech/) serving as the memory layer for the system.**

Prerequisites

You'll need access to a Qdrant instance, which can be set up in one of two ways:

1. Install and run it locally on your machine.
2. Sign up for a free account on Qdrant Cloud.

Either option will provide you with a URL to connect to your instance. The cloud version will also provide an API key for authentication.

You'll need Python 3.10 or higher installed, and we recommend using Poetry for dependency management. Since we'll be working with SaaS models over API, no GPU access is required. Install all necessary libraries with a single command:

poetry install

The project uses different LLMs for different tasks, specifically [Gemini](https://ai.google.dev/) and [Claude](https://www.anthropic.com/api). You'll need to obtain API keys for both services. **While you can swap these models with alternatives in the code, we cannot guarantee the system will maintain the same performance level.** CrewAI makes model switching straightforward, allowing for easy experimentation.

Optionally, you can use [AgentOps](https://agentops.ai/) for observability. This requires signing up for an account and obtaining an API key.

Configuration

Create a `.env` file in the root directory with the following entries:

# Qdrant configuration
QDRANT_LOCATION=http://localhost:6333
QDRANT