AI 产品解读Provides a sidebar and web UI for interacting with locally running AI models (Ollama, Chrome AI/Gemini Nano, OpenAI-compatible APIs) directly from the browser
Users running local AI models lack a convenient, integrated browser interface to chat with their models while browsing. Page Assist bridges this gap by offering a sidebar and full web UI without requiring cloud services or accounts.
User installs extension → runs a local AI model server (e.g., Ollama) → opens the sidebar (Ctrl+Shift+Y) or web UI (Ctrl+Shift+L) → chats with the local model → can also summarize YouTube videos, chat with PDFs, use internet search, or download new models from ollama.com/huggingface.co directly via integrated buttons.
Sidebar chat interface for local AI modelsFull web UI accessible via options pageSupport for Ollama, Chrome AI (Gemini Nano), and OpenAI-compatible APIs (LLaMA.cpp, LM Studio, Llamafile, vLLM)Chat with PDF and documents (PDF, CSV, TXT, MD, DOCX)Internet search integrationYouTube video summarization via content script
- 目标用户
- Developers and power users running local AI models / Privacy-conscious users who prefer local inference over cloud APIs / Ollama users wanting a browser-integrated UI / Users who want to chat with documents and web pages using local models
- Source code is heavily minified/bundled (8.5MB PageAssistProvider chunk), making deep code inspection difficult
- No UI screenshots available to visually confirm feature gating or paid tiers
- The extension's own settings/options UI could not be rendered to check for hidden premium prompts
付费分析未识别到付费功能
No paid features, subscription tiers, or payment flows exist in this extension. The deterministic heuristic found zero paid signals. The extension is open-source (GitHub: https://github.com/n4ze3m/page-assist) and free. All 'pro' keyword hits are false positives from minified library code. No Stripe, billing, premium, upgrade, or payment-related code was found. The extension interfaces with user-hosted local AI models and user-provided API endpoints with no gating of features behind payment.
- 置信度
- 95
- 支付平台
- --
- 来源
- AI / 高
- 需要登录
- 否
- 登录理由
- The extension operates entirely locally with no account system. It connects to user-hosted local AI model servers (e.g., Ollama at 127.0.0.1:11434) and supports user-provided OpenAI-compatible API endpoints. No login, signup, or registration flow was found in any source file. The 'auth' keyword hits in background.js are all from bundled library code (Zod schema validation, JSON Schema), not authentication logic.
Very impressed. I've been using Open WebUI, which is a lot of trouble installing and setting up, only to get super slow LLM web search. Then I found Page Assist and wow, It's so easily set up and the web search is just fast. I only hope the "brain" icon can be changed to something better looking and Page Assist can be saved as a Web APP.
Please add a login/logout feature. I mean, the ability to automatically back up chat history or prompts via email/Gmail authentication, so it can be accessed universally across all browsers at the same time.
One of the simpler web UIs for Local AI
ONE PROBLEM - Web Search is Very Slow and there is need to Add DuckDuckGo API option for AI's search which is also Free for Unlimited Requests and is Fast too, Only This is Needed Otherwise Its best Lightweight Offline AI Extension.
Good!
Couldn't be any easier. Just wish i could specify what drive to store the models.
Top baby
Mi piace molto l'idea, anche se il progetto è allo stato iniziale ma possiede un potenziale enorme. Complimenti continuate così!
Отличная штука
I’ve tried out a bunch of extensions for using local LLMs or custom endpoints, and this one, is by far the best I’ve come across so far. It’s super easy to set up, letting me connect my locally hosted models or custom endpoints without any major headaches. The interface is intuitive and there are tons of customisation options that make it really flexible for both personal and professional projects. I haven’t even started diving into all the custom prompts and things yet, but from what I’ve seen so far, this extension feels great! Hats off to the dev or team that built this. Star it on Github if you have an account!