warpsurf
An open-source AI-powered browser copilot that runs locally in Chrome. Chat, search, or deploy agents to help you surf the web.
warpsurfAn open-source AI-powered browser copilot that runs locally in Chrome. Chat, search, or deploy agents to help you surf the web.
The browser is the door to the internet—it should be open and accessible. Warpsurf provides a model-agnostic AI copilot that can make complex workflows easier while keeping your data local.
For browser automation to be useful, it needs to be fast—enabling "warpsurfing". We're not there yet, but this is the goal. With faster models and parallel execution, we can get there.
Warpsurf has been built as an open source community tool open to contributions and critique. We want to grow the browser automation ecosystem by finding bugs, useful features, and new use cases.
Warpsurf is a research project under active development. Browser automation carries inherent risks: agents may make mistakes and the web can be dangerous. Monitor agents while they work, use capped API keys, and use warpsurf at your own risk.
queries are automatically triaged to the right workflow—chat, search, or agent—for optimal results.
navigates and interacts with any webpage on your behalf, completing complex multi-step tasks.
orchestrate multiple agents in parallel for complex workflows and parallelisable research tasks.
low latency web-augmented chat that finds and summarizes information from across the web.
runs locally in your browser. your data never leaves your machine—your browsing stays with you.
watch agents work in real-time with full transparency. see every action and decision as it happens.
agents operate using their own tab groups, keeping your browsing organized and separate.
real-time token and cost statistics so you always know exactly what you're spending.
incorporate live pricing data from providers for accurate cost predictions.
natural chat interface powered by leading llms for quick questions.
optionally use your browser history to improve agent performance.
preview task duration and cost before the agent starts working.