Zimo AI: A Full-Stack AI - Powered Browser Extension and Web App

Abstract

This paper presents Zimo AI, a full-stack artificial intelligence system designed to assist users while browsing web content. The system consists of a Chrome Browser Extension based on Manifest V3 and a React-based web application powered by Google’s Gemini 2.5 Flash large language model. The objective of the system is to improve user understanding of webpage information by providing intelligent assistance directly within the browsing environment.

The proposed system provides four key functionalities: webpage summarization, question answering based on page content, image analysis, and automatic FAQ generation. The summarization module converts webpage content into structured summaries with key points, while the Q&A module allows users to interact with the page content through natural language queries. The image analysis module generates descriptions and extracts text from images, and the FAQ generator automatically produces relevant question–answer pairs to enhance content comprehension.

The Chrome extension extracts webpage data using a two-stage DOM extraction process and supports a Bring Your Own Key approach for secure API key storage in the browser. The companion web application extends the system with user authentication, persistent conversation history, multi-turn chat, URL-based summarization, and image analysis through file uploads.

The deployed system demonstrates the practical integration of large language model capabilities within a web browsing environment. By embedding AI-powered assistance directly into the browser and web application, the system improves users’ ability to quickly understand webpage content, extract important information, and interact with online resources more effectively. The results indicate that such integration can significantly enhance accessibility, information comprehension, and overall user interaction with web-based content.

Country : India

1 Aditya Ranjan2 Ajay Khoje3 Atharva Dekondwar4 Neil Kashyap5 Prof. Shahrukh Shaikh6 Prof. Mayuri Narudkar

  1. Student, Artificial intelligence and Machine Learning Diploma, Ajeenkya D. Y. Patil School of Engineering, Charholi, Pune, India
  2. Student, Artificial intelligence and Machine Learning Diploma, Ajeenkya D. Y. Patil School of Engineering, Charholi, Pune, India
  3. Student, Artificial intelligence and Machine Learning Diploma, Ajeenkya D. Y. Patil School of Engineering, Charholi, Pune, India
  4. Student, Artificial intelligence and Machine Learning Diploma, Ajeenkya D. Y. Patil School of Engineering, Charholi, Pune, India
  5. Guide, Artificial intelligence and Machine Learning Diploma, Ajeenkya D. Y. Patil School of Engineering, Charholi, Pune, India
  6. HoD, Professor, Artificial Intelligence & Machine Learning Engineering Diploma, Ajeenkya D.Y Patil School of Engineering, Charholi, Pune, India

IRJIET, Volume 10, Issue 3, March 2026 pp. 67-71

doi.org/10.47001/IRJIET/2026.103011

References

  1. M. Lewis et al., "BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation," Proc. ACL, 2020, pp. 7871–7880. [Online]. Available: https://aclanthology.org/2020.acl-main.703
  2. P. Lewis et al., "Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks," Proc. NeurIPS, 2020, pp. 9459–9474. [Online]. Available: https://arxiv.org/abs/2005.11401
  3. A.Radford et al., "Learning Transferable Visual Models from Natural Language Supervision," Proc. ICML, 2021, pp. 8748–8763. [Online]. Available: https://arxiv.org/abs/2103.00020
  4. Google DeepMind, "Gemini: A Family of Highly Capable Multimodal Models," arXiv:2312.11805, 2023. [Online]. Available: https://arxiv.org/abs/2312.11805
  5. Google Chrome Developers, "Chrome Extensions Manifest V3 Overview," 2024. [Online]. Available: https://developer.chrome.com/docs/extensions/mv3/
  6. A.Ranjan, "Zimo AI Chrome Extension Installation Guide," Medium, 2026. [Online]. Available: https://medium.com/@adityavranjan/zimo-ai-installation-guide-c9e6c64fffd8