warpsurf

An open-source AI browser copilot for Chrome. Chat, search the web, or deploy autonomous agents — with voice and multi-agent orchestration.

Add to Chrome Source

Vision

The browser is the door to the internet — it should be open and accessible. Warpsurf provides a model-agnostic AI copilot that makes complex workflows simple while keeping your data local. Bring your own API keys from any supported provider.

Speed

For browser automation to be useful, it needs to be fast — enabling "warpsurfing". With intelligent routing, parallel multi-agent execution, and the fastest models available, we're getting there.

Community

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.

Safety: Warpsurf is a research project under active development. Browser automation carries inherent risks. Use the URL firewall, monitor agents in real-time, use capped API keys, and use warpsurf at your own risk.

Demos

Agent workflow
agent workflow
Summarize page
summarize page
Voice input
voice input
Conversational tools
conversational tools

Features

Workflows
smart router
queries are automatically triaged to the right workflow — chat, search, or agent
autonomous agent
autonomous browsing agent that navigates and interacts with any webpage
multi-agent
orchestrate multiple agents in parallel for complex workflows and research tasks
fast search
low-latency web-augmented chat that finds and summarizes information from across the web
Input
voice input
speech-to-text via OpenAI Whisper or Google Gemini, with optional auto-submit
attachments
drag-and-drop or paste images, text files, PDFs, and JSON
context menus
right-click to explain text, summarize pages, or analyze images
conversational tools
change models, temperature, vision, and context tabs through natural language
Control
take control
agents hand back control at critical junctures for your input
pause & resume
pause running agents with optional instructions when resuming
live follow-up
send instructions to a running agent without stopping it
emergency stop
one-click stop for all running workflows
agent manager
dashboard with search, live tab previews, and one-click task launch
Intelligence
tab context
manually select tabs or enable auto tab context with privacy controls
history context
optionally use browser history with configurable time windows
workflow graph
visual DAG of multi-agent subtasks with real-time status
trajectory view
timeline of agent actions grouped by site
Privacy & Management
privacy first
runs locally with URL firewall for allow/deny domain control
favorites
save, import, and reorder your most-used prompts
session restore
workflows persist and resume seamlessly when you reopen
Cost
usage tracking
real-time token and cost stats per session, role, and model
task estimation
preview steps, duration, and cost before starting
live pricing
live pricing data for accurate cost predictions

Get Started

Chrome Documentation GitHub