browser-use/browser-use
Browser-use is a project that enables AI agents to control and automate web browsers, allowing users to perform complex tasks through natural language commands and AI-driven browser interactions.
Awesome AI Agents › Browser Automation Tools
OSU-NLP-Group/WebDreamer
WebDreamer is a project developed by the OSU-NLP-Group that focuses on enhancing the capabilities of language models (LLMs) as web agents through model-based planning. The core idea is to leverage LLMs as a world model of the internet, enabling them to predict the outcomes of actions on websites and perform speculative planning. This approach addresses the challenge of scaling inference-time compute for language agents, offering a more efficient and flexible alternative to traditional reactive methods and tree search techniques. The project includes a world model module that predicts webpage changes in various formats such as change descriptions, accessibility trees, and HTML structures. This module allows the simulation of multiple steps of interaction with web pages, facilitating better planning and decision-making by the agent. WebDreamer also provides simulation scoring and controller modules to evaluate and select the best actions based on the predicted outcomes. WebDreamer has demonstrated strong performance improvements on benchmarks like VisualWebArena, Online-Mind2Web, and Mind2Web-live, significantly outperforming reactive baselines and showing better efficiency compared to tree search methods. The project releases training data and model checkpoints, including the Dreamer-7B model, which supports image-based observation spaces. The repository is structured to allow independent experimentation with different modules and includes code to reproduce experiments on specific benchmarks. The project emphasizes safety, flexibility, and reduced reliance on real-world interactions by effectively exploring the search space through simulations. Overall, WebDreamer represents a significant advancement in the use of LLMs for web-based tasks, combining model-based planning with large language models to create more capable and efficient web agents.
https://github.com/OSU-NLP-Group/WebDreamer
Also in Browser Automation Tools
CloakHQ/CloakBrowser
CloakBrowser provides a stealthy Chromium binary with C++ source-level fingerprint patches to pass bot detection, serving as a drop-in Playwright/Puppeteer replacement.
jackwener/OpenCLI
OpenCLI converts any website or local tool into a command-line interface, enabling both human interaction and AI agent automation through logged-in browser sessions.
browserbase/stagehand
Stagehand is a production-ready AI web browsing framework that combines code and natural language to enable reliable and extensible browser automation.
browser-use/web-ui
A web-based user interface, powered by Gradio, for interacting with browser automation agents, supporting various LLMs and custom browser configurations.
citrolabs/ego-lite
ego lite is a specialized browser designed for AI agents to perform web automation in parallel with human browsing, offering isolated workspaces and enhanced page snapshots.
jo-inc/camofox-browser
Camofox-browser is a stealth headless browser server, powered by Camoufox (a Firefox fork), designed for AI agents to bypass bot detection and Cloudflare, offering a Puppeteer/Playwright alternative.
microsoft/fara
Fara-7B is a compact and efficient agentic small language model by Microsoft designed to perform multi-step computer tasks through visual interaction, achieving state-of-the-art performance on web automation benchmarks while enabling on-device deployment for privacy and low latency.