Prompt Paul- AI Insights at the click of a mouse

About

✨PromptPaul: A sidebar for your LLM✨
Most browser AI extensions are either tied to a single subscription or filled with clutter you don't need. PromptPaul is a straightforward sidebar that connects to the LLM APIs you already use—Mistral, OpenAI, Anthropic, or a local vLLM. It handles the extraction so you can focus on the conversation.

➼ It knows what you're looking at...
When you click a preset, PromptPaul identifies the page type. For a YouTube video, it pulls the transcript and speaker notes. On GitHub, it looks for code structure and dependencies. For articles, it focuses on the facts. It’s better than a generic "summarize" command because the prompt changes based on the context.

➼ Visuals for YouTube...
Summaries can be disorienting if you can't see what the speaker is pointing at. 
PromptPaul takes screenshots of the video at key moments while it processes the transcript, providing a set of visual "slides" in the chat alongside the text.

➼ Beyond just text...
👉 Images and Media: Right-click any image, video, or audio file on the web to send it directly to your model's vision or multi-modal API.
👉 Voice Input: Use the built-in microphone button to dictate prompts when you’re tired of typing.

➼ Profiles and Presets...
You can save different API configurations as profiles and pin your frequent instructions as quick-action buttons. It’s designed to stay out of the way until you need it, using JetBrains Mono for a clean, readable interface that feels like a dev tool rather than a marketing widget.

It's just a better way to use an AI while you browse......without losing your flow.