sleepinginsummer

sleepinginsummer / agent-browser-cli

Public

使用 agent-browser-cli 进行浏览器感知与控制。适用于标签页扫描/切换、页面 JS 执行、Cookie、CDP、contentSettings、截图、文件上传、下拉框点击、tmwd_cdp_bridge 初始化和 Web 工具排障

16
4
69% credibility
Found May 13, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

A tool that bridges AI agents to a user's existing web browser session for reading pages, executing actions, and reusing authentication without starting fresh.

How It Works

1
🔍 Discover the Tool

You hear about a handy helper that lets smart assistants use your real web browser, keeping all your logins and open pages just as they are.

2
📥 Set It Up Quickly

Follow a simple guide to download and prepare the tool on your computer, taking just a minute or two.

3
🧩 Add Browser Helper

Load a small companion piece into your web browser so it can share what's on your screen with the assistant.

4
🚀 Start the Connection

Run the tool once to link it to your browser, and it stays ready in the background for whenever you need it.

5
👀 See Your Pages

Ask it to list or peek at your open web pages, and get a clear snapshot of text, links, or the whole view instantly.

6
✨ Let Assistant Act

Have your smart assistant click buttons, fill forms, or read details on any page, feeling just like you're doing it yourself.

✅ Tasks Flow Smoothly

Now your assistant handles web chores with your personal logins preserved, saving you time and hassle every day.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 16 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is agent-browser-cli?

Agent-browser-cli is a Rust-powered CLI on agent browser cli github that hooks into your existing Chrome browser via a lightweight extension, letting you scan tabs, execute JS, grab cookies, control CDP sessions, take screenshots, upload files, and click dropdowns without spinning up headless browsers. It turns real Chrome sessions—complete with logins and state—into reusable web automation tools for agents, with commands like `agent-browser-cli tabs`, `scan`, `exec`, and `open`. JavaScript runs in the page context for low-latency browser client agent ops, beating traditional setups for quick web tasks.

Why is it gaining traction?

Unlike Playwright-cli or agent-browser vs Playwright, which launch fresh instances and lose cookies, this reuses your live Chrome tabs for instant access to authenticated sites, delivering 50ms reads and 300ms complex queries. Agent browser cli skill integration shines for AI workflows, with agent-browser-cli vercel deploys possible via npm, and cline browser agent hooks like cursor agent cli browser make it a fast alternative for web agent browser click automation. The persistent daemon avoids per-call overhead, perfect for high-frequency agent calls.

Who should use this?

AI agent builders chaining LLMs like Claude or Codex with browser actions, especially for scraping logged-in dashboards or form interactions. Web devs prototyping agent-browser-cli skills for tasks like tab switching or conditional JS waits. Automation scripters ditching heavy frameworks for lightweight CLI browser agent control in JS-heavy pipelines.

Verdict

Grab it if you're building browser-aware agents—solid for niche reuse of real sessions, with clear docs and skill templates. At 16 stars and 0.7% credibility, it's early but MIT-licensed and performant; watch for more tests before production.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.