PageTalk: An Open-Source Browser AI Extension 78

小编 ·

PageTalk: An Open-Source Browser AI Extension

PageTalk is an open-source browser AI extension powered by Google's Gemini API to enhance web browsing. It lets users summarize web content, chat with context, and easily manage custom AI agents.

PageTalk: An Open-Source Browser AI Extension

PageTalk is an open-source browser AI extension that leverages Google's Gemini API to enhance the web browsing experience. Users can summarize web page content, engage in context-aware chats, and easily manage custom AI agents.

PageTalk: An Open-Source Browser AI Extension.webp

PageTalk Main Features

  • Web Page Interaction: Reads web page content for contextual conversations.

  • Gemini API Integration: Leverages Gemini for powerful language processing tasks.

  • Contextual Chat: Discuss the current web page content with AI.

  • Multi-Agent System: Create, customize, switch, and import/export AI agents.

  • Model Selection: Supports choosing from multiple Gemini models.

  • Image Input: Supports uploading or pasting images for discussion.

  • Rich Content Rendering: Supports rendering of Markdown, code highlighting, LaTeX, and Mermaid diagrams.

  • Personalization: Configurable API key, language (English/Chinese), and theme (light/dark).

  • Chat Export: Save chat history as Markdown or text files.

  • Adjustable Panel: Adjust the width of the sidebar panel.

PageTalk Installation Steps

  1. Obtain API Key: Before using the extension, you need to get a Gemini API key from Google AI Studio.

  2. Visit its open-source address: https://github.com/jeanchristophe13v/PageTalk.

  3. Download the latest version of the compressed package.

  4. Edge: Open edge://extensions/, enable "Developer mode", click "Load unpacked extension", and select the project folder.

  5. Chrome: Open chrome://extensions/, enable "Developer mode", click "Load unpacked extension", and select the project folder.

PageTalk: An Open-Source Browser AI Extension.webp

How to Use PageTalk

  • Open the Extension: Click the PageTalk icon or use the shortcut (default Alt+P).

  • Chat Tab:

  • Select a model/agent.

  • Extract page content (automatic or manual); the status shows context length.

  • Enter a message or paste/upload an image.

  • Send the message (press the Enter key or click the button).

  • Use the "Summarize" quick action.

  • Clear history/context (click the trash can icon).

  • Hover over a message to copy/delete/regenerate.

  • LaTeX ($...$, $$...$$) or Mermaid (mermaid ... ) syntax in AI responses will be automatically rendered. Click on a Mermaid diagram to zoom/pan.

  • Settings Tab: Includes General, Agents, and Models sub-tabs.

  • General: Switch language/theme, export chat history.

  • Agents: Manage agents, import/export configurations.

  • Models: Set API key, select default model, toggle auto-extraction.

Use Cases

  • Learning & Research: Helps users quickly summarize web page content and grasp key information.

  • Information Retrieval: More efficiently obtain information from web pages through conversations with the AI assistant.

  • Multimodal Interaction: Supports image input for a richer interactive experience.

Github:https://github.com/jeanchristophe13v/PageTalk