Zot gets a Chrome extension for browser control

A new Chrome extension and local bridge let zot, a terminal-based coding agent, inspect and control browser tabs through a side panel. The project installs as a zot extension, then adds a zot-chrome command shim and connects to Chrome over a local WebSocket.

Zot gets a Chrome extension for browser control

A new Chrome extension for zot lets the terminal coding agent operate browser tabs from a Chrome side panel, according to the project’s README on GitHub. The setup pairs a local bridge with a browser extension so zot can inspect pages, click buttons, type into fields, and handle common browser tasks without changing zot itself.

The project is called zot-chrome-operator and is installed as a zot extension first. The maintainer says the install command copies the extension files, but does not run installation hooks, which is why a second command is needed to create a global zot-chrome command shim.

⚡ New to this?

This is a tool that connects a coding agent to a real browser, so the agent can do web tasks instead of only reasoning in a terminal. A browser extension is the add-on that runs inside Chrome, and a local bridge is the background process that lets the agent send and receive browser actions. For people who automate work or test web apps, that combination matters because it turns a text-based agent into something that can interact with pages directly.

🦞 OpenClaw angle

If you build self-hosted agents, treat this as a pattern for adding browser control without modifying the core agent. Keep the browser bridge local, use a fixed localhost port, and make the extension reload flow part of your deployment checklist so updates do not break automation. Also note the no-session design: if you need persistent workflows, you will have to store state outside the bridge rather than assuming the agent remembers past runs.

The install process starts with:

zot ext install https://github.com/patriceckhart/zot-chrome-operator node "$HOME/Library/Application Support/zot/extensions/zot-chrome-operator/bin/install-cli.js"

That script creates the zot-chrome command at ~/.local/bin/zot-chrome. The README says users should make sure ~/.local/bin is on their PATH, so the command can be run from any directory.

Once installed, users can check the setup with zot-chrome status. They can also build the unpacked Chrome extension by running zot-chrome ext. On the first run, the command installs npm dependencies and builds the extension if needed, then prints the path to a dist directory that can be loaded into Chrome.

To finish the browser setup, users open chrome://extensions, turn on Developer mode, click Load unpacked, and select the printed dist directory. The README also says users can pin the zot extension for quick access and open the side panel from the extension icon.

The bridge itself starts with zot-chrome start. According to the README, that command ensures the extension is built, then launches a local bridge in the background. The bridge starts zot rpc --no-session with the bundled zot extension, and the Chrome side panel connects to ws://localhost:9224.

The tool can list, create, switch, and close tabs. It can also inspect page context, navigate, click, type into native and rich text editors, select options, scroll, extract page text, and wait.

The project exposes a few more commands for maintenance. zot-chrome stop stops the bridge, zot-chrome logs shows bridge logs, and zot-chrome status checks whether it is running. The README says users can set ZOT_PROVIDER=anthropic and ZOT_MODEL=claude-sonnet-4-5 before starting the bridge if they want to use those settings, and PORT=9225 is shown as an example for changing the port.

The author says no chat session is persisted, because the bridge uses zot --no-session. The project is released under the MIT license.

Source: HN Show HN ↗

More from OpenClaw News