yohasebe/openai-chat-api-workflow

What it solves

This project provides a seamless way to integrate OpenAI's GPT models directly into the macOS Alfred 5 productivity tool. It eliminates the need to switch between different applications or browser tabs by bringing chat, image generation, and file analysis capabilities directly to the desktop environment.

How it works

The workflow acts as a interface between the user's Mac and the OpenAI API. It allows users to interact with AI models via three primary methods: Alfred's command line, selected text in any application, or a locally hosted web UI. It supports a wide range of OpenAI's capabilities, including text generation, image creation and iterative refinement, file understanding (PDFs, Office docs, code), and speech-to-text/text-to-speech synthesis.

Who it’s for

Mac users who use Alfred 5 (with Powerpack) and have an OpenAI API key, seeking to integrate AI-powered productivity tools into their system-wide workflow.

Highlights

• Multimodal Interaction: Supports text, images, PDFs, Office documents, and code analysis.
• Integrated Web UI: A local web interface for managing conversations and uploading files.
• Iterative Image Editing: Ability to refine generated images through follow-up prompts.
• Speech Capabilities: Integrated speech-to-text (transcription) and text-to-speech (synthesis).
• Flexible Execution: Trigger AI queries via Alfred keywords, hotkeys, or by sending selected text.
• Conversation Management: Maintains history on a single scrollable page with JSON export/import functionality.