https://github.com/user-attachments/assets/d5407fc3-be1f-4b1d-8f88-afb7468ec917

# Browsentic - a completely integrated agentic browser harness

**No account · No auth · No API key · No cloud · No telemetry · Local first** Your real, logged-in browser, driven from an AI side panel by the agent CLI you already have.
Also an optional [MCP server](docs/guide/mcp-clients.md), so any MCP client can drive the same browser.

Works in

Runs on

## Quick Start Browsentic is your browser's superpower, free and open source. One install sets up everything, the extension included. **macOS**: one line installs [the app](docs/guide/mac-app.md), which brings Node and everything else: ```sh curl -fsSL https://browsentic.com/install.sh | sh ``` **Windows**: one line in PowerShell installs [the Windows app](docs/guide/windows-app.md) ([experimental](docs/guide/limits.md#windows-is-experimental)): ```powershell irm https://browsentic.com/install.ps1 | iex ``` **Any platform**, with [Node.js](https://nodejs.org) 20 or newer: ```sh npx browsentic@latest setup ``` Setup asks which browser you use and opens Browsentic's store page in it (the apps do the same from their Overview tab). Press **Add to Chrome** (**Add** in Firefox; Edge first asks you to allow extensions from other stores), and the extension connects to Browsentic Bridge by itself; if it asks for a code, enter the one setup shows. From then on it updates itself from its store. The extension is on the Chrome Web Store for Chrome, Brave, Arc, Vivaldi and Opera, on Edge Add-ons for Edge, and Firefox gets an add-on signed by Mozilla:

If you added the extension from a store first, run the same line, and the two connect as soon as the Bridge is in place. The side panel runs on an agent CLI you are signed in to: [Claude Code](https://claude.com/claude-code), [Codex](https://developers.openai.com/codex/cli) or [another supported one](docs/guide/agents.md). It uses the one you already have, so there is no API key, no account and no extra subscription. Full guide, Firefox and unpacked installs included: **[browsentic.com/docs/guide/install](https://browsentic.com/docs/guide/install/)** ## Key Capabilities

🧭 Side panel, human in the loop	🤖 Every major agent CLI	🎯 50+ deterministic tools
🌐 All Chromium + Firefox	🗺️ Full browser navigation	🧩 Solves captchas on the way
📄 Reads files, acts on them	↕️ Uploads & downloads	🧠 Creates its own skills
⚡ Writes & runs live tools	🎙️ Voice, text or show it once	👁️ Point at what you mean
⏱️ Monitors & schedules	🛡️ Guardrails & approvals	🔌 Optional MCP server
📱 Drives Chrome on your Android phone (experimental)

## How It Works ![How an instruction becomes a click](https://browsentic.com/flow.png)

The same flow, as text

``` You ──speak or type──> Extension ──local WebSocket──> Browsentic Bridge ──spawns──> your agent CLI ▲ (claude │ codex │ agy │ …) └──────────────────── page actions ──────────────────────┘ Any MCP client ──stdio──> browsentic mcp ──> the same Bridge ──> the same browser ```

The extension dials out to Browsentic Bridge, because a Manifest V3 service worker cannot listen for connections. One Bridge owns the browser link, so several MCP clients can share one browser. Everything binds to `127.0.0.1`. ## Side Panel and Desktop Apps ![The side panel in Chrome partway through a run, waiting for approval to send a reminder; the Mac app's Overview and its one-click setup; the Windows app's agent picker and its Android phone checks](docs/assets/side-panel-and-apps.webp) The [side panel](docs/guide/features/conversations.md) is where you work with the agent. Every page action shows up as a step, and consequential ones, such as submitting a form, wait for your [approval](docs/guide/approvals.md). A new tab opens [Browsentic Home](docs/guide/features/new-tab.md), a full-page composer with your frequent sites, which hands the task to the side panel as soon as you send it; you can switch it off. The desktop apps for [macOS](docs/guide/mac-app.md) and [Windows](docs/guide/windows-app.md) ([experimental](docs/guide/limits.md#windows-is-experimental)) install Node and Browsentic Bridge in one click, turn the Bridge on and off, add the extension to each browser, pick the [agent CLI](docs/guide/agents.md) and its model, and check an [Android phone](docs/guide/features/android.md) before the side panel drives it. ## Resources - 📚 [Documentation](docs/) - 🚀 [Install and Pair](docs/guide/install.md) - ✨ [Features](docs/guide/features/) - 🧰 [All 52 Page Tools](docs/reference/tools.md) - 🔌 [Optional: Drive It From Claude Code, Cursor or Zed Over MCP](docs/guide/mcp-clients.md) - 🛡️ [Approvals and Guardrails](docs/guide/approvals.md) - 🏗️ [Architecture](docs/internals/) - 🩺 [Troubleshooting](docs/guide/troubleshooting.md) ## Privacy and Security Nothing connects except Browsentic's own extension, paired by itself through the browser or with a code, and both ends prove who they are. The agent asks before any consequential action, credentials on a page are sealed before the agent sees them, and the sites you [block](docs/guide/features/blocked-sites.md) are off-limits to every agent: the browser enforces that list, where no agent can reach it. The full model is in [SECURITY.md](SECURITY.md), and what it does not cover is in [Limits](docs/guide/limits.md). ## Contributing Bugs and ideas are welcome. Start at [CONTRIBUTING.md](CONTRIBUTING.md). ## License Browsentic is licensed under the [Apache License 2.0](LICENSE). Versions up to 0.7.15 were released under the MIT License. - **Source Available**: Always visible source code - **Local First**: No cloud component, no telemetry, no account - **Extensible**: Add your own page capabilities, skills and agent runners ## What does Browsentic mean? **Short answer:** "Browse" + "agentic". **Long answer:** Most browser automation hands the work to a *different* browser: a headless one, in a container, logged in to nothing. Browsentic works the other way round. The agent acts in the browser you are already looking at, with the sessions you are already signed in to: browsing, made agentic, where you already browse.