Support
A Kimi Help Center page covers troubleshooting, and the homepage FAQ explains connection fixes such as re-sending the command and restarting Kimi Desktop.
Learn moreA browser extension that lets AI agents click, type and read the web

Kimi Browser Extension is a productivity tool. Kimi Browser Extension (formerly Kimi WebBridge) lets AI agents operate a real Chrome or Edge browser locally: opening pages, clicking, filling forms and extracting data, with a Kimi sidebar and web operation recording.
Kimi Browser Extension, previously named Kimi WebBridge, gives an AI agent hands inside a real browser. Instead of only answering questions about the web, the agent can open pages, press buttons, fill in forms and pull information off the screen, covering the repetitive work that normally falls to a person with a mouse and a dozen open tabs. The product page points to jobs such as trend research, job hunting and flight comparison, and lists automated reporting, intelligent form-filling and quantitative backtesting as common use cases. The extension works by pairing itself with a local service on the user's device. When Kimi sends a request, that service uses the Chrome DevTools Protocol to drive the Chrome or Edge window the user already has open. It can move between pages, click, take screenshots and read page content, then return the results to Kimi. Because the process runs locally, login sessions and browsing content are described as staying on the machine, which matters for sites that sit behind an account. There are several ways in. The sidebar entry is opened from the Kimi icon in the browser toolbar and requires signing in with a Kimi account. Kimi Desktop and other agents can also call the extension through a local agent, which works as it did under the old WebBridge name. The rename brought new features: a browser sidebar, recording of web operations, and a way to break web pages down into commands. Installation is through the Chrome Web Store, with a manual install route for people who cannot reach the store. Among alternatives, it fits between generic browser automation scripts and standalone cloud agents. Compared with script-based tools it asks for no code, and compared with cloud browsers it reuses the user's own signed-in browser. Its tight link to the Kimi ecosystem is both the draw and the limit, since the sidebar depends on a Kimi account and the best experience assumes Kimi's own agents.


A Kimi Help Center page covers troubleshooting, and the homepage FAQ explains connection fixes such as re-sending the command and restarting Kimi Desktop.
Learn moreWorks with Chrome and Edge, Kimi Desktop and other local agents that call the extension. Uses a Kimi account for the sidebar.
Opens pages, clicks buttons, fills forms and extracts page content, takes screenshots, and drives Chrome or Edge through the DevTools Protocol. Includes a browser sidebar, web operation recording and splitting pages into commands.





