AI Agent: OpenAI Capabilities Showcase & Demo Toolkit

Unlock the Full Spectrum of OpenAI's Power with this AI Agent Toolkit

This n8n workflow serves as a comprehensive demonstration and testing ground for a wide array of OpenAI's AI models and functionalities. It's designed for AI natives to quickly understand and experiment with different use cases. This AI Agent acts as your personal R&D lab for OpenAI, showcasing a suite of powerful abilities:

• AI-driven Text Summarization: Condense long texts into concise Tl;dr summaries using models like davinci-003 and gpt-3.5-turbo.
• Multilingual Translation: Translate text between languages efficiently.
• Creative Text Generation: Draft email replies, generate marketing copy, or brainstorm ideas with customizable system prompts.
• DALL-E Image Generation: Create unique visuals and illustrations directly from text descriptions, including examples of using ChatGPT to refine image prompts.
• HTML/SVG Code Generation: Automate the creation of simple web graphics or code structures using AI.
• Audio-to-Text Transcription (Whisper): Includes a blueprint (disabled by default) for converting spoken words into text using OpenAI's Whisper model via direct API call.

Key Features & Benefits of this Toolkit

• Hands-On OpenAI Exploration: Interactively test various models (davinci-003, gpt-3.5-turbo, DALL-E, Whisper) and their applications.
• Side-by-Side Demonstrations: See different prompting techniques, model behaviors (e.g., completion vs. chat, edit operations), and output formats in one place.
• Learn by Example: Understand how to integrate OpenAI into n8n for tasks like summarization, translation, image creation, and code generation.
• Programmatic API Control: Includes an example of calling the OpenAI Chat Completions API directly via the HTTP Request node for full customization and understanding message array structures.
• Image Generation Workflow: Illustrates chaining ChatGPT for prompt engineering with DALL-E for refined image output.
• Foundation for Your Agents: Use these examples as building blocks to create your own specialized AI agents for specific business needs.
• Clear Explanations: Sticky notes throughout the workflow guide you through each example and offer best practice advice (e.g., model cost considerations).

• An n8n instance (Cloud or self-hosted).
• OpenAI API Key with access to models like gpt-3.5-turbo, davinci-003, DALL-E, and Whisper (for the audio example).
• (Optional) An audio file in a supported format (e.g., MP3) if you wish to test the Whisper transcription example.

1. Download the n8n workflow JSON file.
2. Import the workflow into your n8n instance.
3. Configure your OpenAI API credentials. For all 'OpenAI' nodes and the 'Whisper-transcribe' HTTP Request node, select your pre-configured OpenAI credential in n8n.
4. Review the 'Text-example' Code node. This provides the sample text input for many branches. Modify this text to experiment with your own content.
5. The workflow is manually triggered. To test specific functionalities, you can execute individual branches by selecting the last node of a branch and clicking 'Execute Node', or run the entire workflow (be mindful of API call volume).
6. For the audio transcription example ('LoadMP3' and 'Whisper-transcribe' nodes): a. These nodes are disabled by default. Enable them to test. b. In the 'LoadMP3' node, update the fileSelector parameter to the correct path of your audio file if self-hosting n8n with local file access. Alternatively, adapt it to receive binary data from a previous step (e.g., HTTP download). c. Ensure the 'Whisper-transcribe' HTTP Request node is correctly configured with your OpenAI credentials.
7. Explore each branch. The sticky notes provide context for each example. Modify prompts within the OpenAI nodes to tailor experiments to your needs.
8. This workflow is primarily a demonstration toolkit. Adapt and integrate these patterns into your own production AI Agents.