Skip to main content
BrowserWire watches you browse a site and builds a versioned manifest of everything the page can do — entities, actions, views, and workflows — served as a REST API your AI agent already knows how to call. 
You browse a site → BrowserWire watches → You get a typed API

Quick Start

Install BrowserWire and discover your first site in minutes

API Reference

Full REST API reference with request and response examples

LLM Providers

Connect OpenAI, Anthropic, Gemini, or Ollama

Agent Integration

Feed discovered APIs to your AI agent

​
How it works

1

Install & configure

Install the BrowserWire CLI and set your LLM provider API key.
2

Load the Chrome extension

Load the bundled extension in developer mode to capture page interactions.
3

Explore a site

Open any website, start an exploration session, and navigate around. BrowserWire watches what’s on the page and what you can do with it.
4

Call the API

Get an OpenAPI spec for the discovered site. Point your agent at it — or call actions and workflows directly over REST.

Discover a Site

Step-by-step walkthrough of an exploration session

Desktop App

Use the native desktop app for an integrated experience

Core Concepts

Understand manifests, entities, actions, and workflows

CLI Reference

All CLI flags and options