What is Webinar Openai Agents Mcp

webinar-openai-agents-mcp is a project that demonstrates how to use the Model Context Protocol (MCP) to orchestrate AI agents utilizing OpenAI SDK, Augment Code, and Qdrant for enhanced documentation processing.

Use cases

Use cases include automating the retrieval of documentation for Python packages, enhancing code indexing and searching capabilities, and orchestrating AI agents for various documentation-related tasks.

How to use

To use webinar-openai-agents-mcp, clone the repository, install dependencies with Poetry, configure the environment variables in a .env file, and run the main script to process documentation queries.

Key features

Key features include building a pipeline of AI agents for documentation processing, utilizing OpenAI’s Agent SDK for natural language processing, integrating Qdrant’s MCP server for vector search capabilities, and storing/retrieving code snippets with semantic context.

Where to use

webinar-openai-agents-mcp can be used in fields such as software development, documentation management, and AI research, where natural language processing and efficient data retrieval are essential.

Webinar: Using MCP to Orchestrate AI Agents

This repository contains the code for the webinar demonstrating how to use Model Context Protocol (MCP) to orchestrate AI Agents with OpenAI SDK, Augment Code, and Qdrant.

[!NOTE]
If you would like to see the Django application created during the webinar, please check the
simple-django-app directory.

Overview

The project showcases how to:

• Build a pipeline of AI agents for processing documentation
• Use OpenAI’s Agent SDK for natural language processing
• Integrate Qdrant’s MCP server for vector search capabilities
• Store and retrieve code snippets with semantic context through the mcp-server-qdrant

Project Structure

.
├── src/
│   └── code-indexing-agents/
│       ├── main.py           # Main application logic
│       ├── models.py         # Pydantic models
│       └── helper.py         # Utility functions
├── poetry.lock              # Poetry dependencies lock file
├── pyproject.toml          # Project configuration
└── README.md               # This file

Prerequisites

• Python 3.10 or higher
• Poetry package manager
• Qdrant instance (cloud or self-hosted)
• OpenAI API key

Installation

1. Clone the repository:

   git clone https://github.com/qdrant/webinar-openai-agents-mcp.git
   cd webinar-openai-agents-mcp
   

2. Install dependencies using Poetry:

   poetry install
   

3. Create a .env file with your configuration:

   QDRANT_URL=<your-qdrant-url>
   QDRANT_API_KEY=<your-qdrant-api-key>
   COLLECTION_NAME=<your-collection-name>
   TOOL_STORE_DESCRIPTION=<description-for-store-tool>
   TOOL_FIND_DESCRIPTION=<description-for-find-tool>
   

Usage

Run the main script to collect and process documentation:

poetry run python src/code-indexing-agents/main.py

The script will use the input query to find the documentation of a Python package, parse it, and store the code
snippets in Qdrant through the MCP server.

[!NOTE]
Please modify the main.py file to pass a different request. By default, we load the docs of django-semantic-search

License

Apache 2.0 License - see the LICENSE file for details.