AI Browser Action Layer

Let AI use your logged-in browser to read common sites cheaply

Problem

Existing AI browser control tools handle logged-in sites like Bilibili, Zhihu, and Xiaohongshu poorly: reading pages back and forth, failing to find content, being very slow, and blowing up tokens with messy design; one developer ended up hand-writing extraction logic for 30+ sites.

Solution

A browser extension plus local CLI that wraps common sites into structured commands, extracts data with in-page JS to avoid dumping full HTML into context, and supports functional composition and piping to turn multi-step browsing into reusable scripts.

Users

Developers and power users who want AI to research, compare prices, and collect content automatically, plus personal automation builders.

MVP

A browser extension that keeps login state and executes commands; a local CLI with site commands and composition syntax; about 10 high-frequency site adapters; structured JSON output for downstream LLMs or scripts; a record/debug mode for adding new sites.

Revenue

Open-source core with paid or subscription site adapter marketplace; enterprise tier for private site adapters and team-shared scripts.

Why now

Agent adoption is bottlenecked by login state and web reading cost, and a browser extension is the lightest breakthrough; a solo dev can validate with 10 sites and let the community expand.