Would it be possible to consider supporting image export?

[qianjun2659]

Currently, page-assist supports text and markdown formats. However, in some scenarios, it is necessary to share the Q&A results in chat tools. Simply copying text or markdown may result in the loss of some markdown features, such as formulas, tables, etc.

Therefore, I would like to inquire if it is possible to consider supporting image-based output for the results. The style could be similar to the following:

Question: What are the features of page-assist?

Vendor-Model:
**Page Assist** is a Chrome extension with the following main features:

1. **Sidebar**:
   - Allows users to open a sidebar on any webpage, enabling interaction with the model and viewing results.

2. **Web UI**:
   - Provides a web interface similar to the ChatGPT website, allowing users to interact with the model.

3. **Chat With Webpage**:
   - Users can chat with the content of a webpage and ask questions about the webpage content.

Additionally, Page Assist supports manual installation, requiring the following prerequisites:
- **Bun** (a JavaScript runtime, [Installation Guide](https://bun.sh/))
- **Ollama** (a local AI provider, [Installation Guide](https://ollama.com/))
- Any endpoint compatible with the OpenAI API (such as LM Studio, llamafile, etc.).

If you have more feature requests, you can suggest them to the developers by creating an issue.

You can also watch a demonstration of Page Assist on [YouTube](https://www.youtube.com/watch?v=8VTjlLGXA4s).

----
BY: page-assist

This approach would help resolve the issue of markdown or text distortion in chat tools while also promoting the tool.

[n4ze3m]

Oh, it's possible. I guess I will look into it. :)

[qianjun2659]

I'm glad to see the image export feature, and I have a couple of small suggestions:

1. When exporting chat images, currently, they are exported in Markdown text format. Could it be possible to export them in a WYSIWYG (What You See Is What You Get) format after Markdown conversion?

2. When exporting images, could the generated image names be slightly adjusted? For example, naming them as "PageAssist-chat-[timestamp].png" or "PageAssist-chat-[chat title].png". This would help avoid generating many duplicate files with the same name each time they are exported.

[n4ze3m]

yes, I will improve it further in the coming updates. That's why I haven't closed this issue yet.