GuideMe & Viewer Integration: Streamline Workflow Results

Hey guys! Ever felt like sifting through your workflow results after running GuideMe is a bit of a hassle? You're not alone! We've got an awesome enhancement in the works to make your life easier. This article dives deep into how we're integrating the Viewer directly into the GuideMe workflow, so you can access your results with just a click. Get ready to boost your productivity and streamline your image description process!

The Problem: Viewing Results Can Be a Pain

Currently, GuideMe helps you through a series of steps, from choosing your AI provider (like Ollama, OpenAI, or Claude) to selecting your model and configuring API keys. You set your input directory, workflow name, prompt style, and finally, launch the workflow. But then what? There's no direct, easy way to view the results once the process is complete. You're left digging through directories, which, let's be honest, isn't the most fun way to spend your time.

GuideMe's Current Workflow Steps:

To really understand the problem, let's break down the current workflow:

1. Choose AI provider: You get to pick your favorite AI engine, whether it's the open-source prowess of Ollama, the established power of OpenAI, or the sophisticated capabilities of Claude.
2. Select model: Tailor your AI even further by choosing the specific model that best suits your needs. This allows for fine-grained control over the output.
3. Configure API key: If needed, you'll set up your API key to ensure seamless communication with your chosen AI provider. This is a crucial step for accessing the AI's functionalities.
4. Choose input directory: Point GuideMe to the folder containing the images or data you want to process. This is the starting point for your workflow.
5. Set workflow name (optional): Give your workflow a descriptive name to easily track and manage your projects. This helps you stay organized, especially when running multiple workflows.
6. Select prompt style: Customize the way GuideMe interacts with the AI by selecting a specific prompt style. This allows you to influence the tone and content of the generated descriptions.
7. Launch workflow: Finally, kick off the magic and let GuideMe do its thing! This is where the AI processes your input and generates the descriptions.

But here's the missing piece: there's no straightforward way to view the results after the workflow finishes. This is where our proposed enhancement comes in, bridging the gap between processing and reviewing your output.

The Power of the Viewer

The viewer (viewer/viewer.py), on the other hand, is a powerful tool. It lets you:

• Open directories via command line: Just type viewer.exe <directory_path> and you're in!
• Browse completed workflows in HTML mode: See your results in a clean, organized HTML format.
• Monitor in-progress workflows in Live mode: Watch the magic happen in real-time as your descriptions are generated.
• Redescribe images with Ollama: Tweak and refine your descriptions using the power of Ollama.

This nifty tool, located at viewer\viewer.exe (once built), is a game-changer for anyone working with image descriptions. But the current disconnect between GuideMe and the Viewer means you have to manually navigate to the output directory and launch the viewer yourself. It's a bit clunky, and we can definitely make it smoother!

The Solution: A Seamless