ko-browser

Browser 自动化 with ko-browser

The 命令行工具 uses Chrome/Chromium via CDP directly. 安装 via go 安装 github.com/libi/ko-browser/cmd/kbr@latest or build from source. 运行 kbr 安装 to 验证 Chrome is avAIlable, or kbr 安装 --with-deps to auto-安装 it. Existing Chrome, Brave, and Chromium 安装ations are 检测ed automatically.

安装ation # 安装 kbr binary directly (no CGO, no external dependencies) go 安装 github.com/libi/ko-browser/cmd/kbr@latest

# Or build from source git clone https://github.com/libi/ko-browser.git cd ko-browser go build -o kbr ./cmd/kbr/ mv kbr /usr/local/bin/

# 验证 browser dependency kbr 安装

# Auto-安装 Chrome if missing kbr 安装 --with-deps

OCR is optional. The default 安装 has zero CGO dependencies. To enable kbr snapshot --ocr, rebuild with -tags=ocr (requires Tesseract):

# 安装 Tesseract first: brew 安装 tesseract (macOS) / apt 安装 libtesseract-dev (Linux) CGO_ENABLED=1 go 安装 -tags=ocr github.com/libi/ko-browser/cmd/kbr@latest

Manual browser 安装ation by OS:

# macOS brew 安装 --cask google-chrome

# Linux (Debian/Ubuntu) sudo apt-获取 安装 -y chromium-browser

# Linux (Alpine) apk 添加 chromium

# Windows choco 安装 googlechrome

Or 下载 from: https://www.google.com/chrome/

Core 工作流

Every browser 自动化 follows this pattern:

Navigate: kbr open Snapshot: kbr snapshot -i (获取 element IDs like 1, 2, 3) Interact: Use numeric IDs to 命令行工具ck, fill, select Re-snapshot: After navigation or DOM changes, 获取 fresh IDs kbr open https://example.com/form kbr snapshot -i # 输出: # 1: textbox "EmAIl" # 2: textbox "Password" # 3: button "Submit"

kbr fill 1 "user@example.com" kbr fill 2 "password123" kbr 命令行工具ck 3 kbr wAIt load kbr snapshot -i # 检查 结果

Command ChAIning

Commands can be chAIned with && in a single shell invocation. The browser persists between commands via a background daemon, so chAIning is safe and more efficient than separate calls.

# ChAIn open + wAIt + snapshot in one call kbr open https://example.com && kbr wAIt load && kbr snapshot -i

# ChAIn multiple interactions kbr fill 1 "user@example.com" && kbr fill 2 "password123" && kbr 命令行工具ck 3

# Navigate and capture kbr open https://example.com && kbr wAIt load && kbr screenshot page.png

When to chAIn: Use && when you don't need to read the 输出 of an intermediate command before proceeding (e.g., open + wAIt + screenshot). 运行 commands separately when you need to 解析 the 输出 first (e.g., snapshot to discover IDs, then interact using those IDs).

Handling Authentication

When automating a site that requires 记录in, choose the 应用roach that fits:

Option 1: Persistent 性能分析 (simplest for recurring tasks)

# First 运行: 记录in manually or via 自动化 kbr --性能分析 ~/.my应用 open https://应用.example.com/记录in # ... fill 凭证s, submit ...

# All future 运行s: already 认证d kbr --性能分析 ~/.my应用 open https://应用.example.com/仪表盘

Option 2: 状态 file (manual save/load)

# After 记录ging in: kbr 状态 导出 ./auth.json

# In a future 会话: kbr 状态 导入 ./auth.json kbr open https://应用.example.com/仪表盘

Option 3: Auth vault (凭证s stored, 记录in by name)

kbr auth save mysite --url https://应用.example.com/记录in --username user --password pass kbr auth 记录in mysite

auth 记录in navigates to the URL and wAIts for 记录in form selectors to 应用ear before filling/命令行工具cking.

Option 4: 会话 name (isolate parallel tasks)

kbr --会话 my应用 open https://应用.example.com/记录in # ... 记录in flow ... kbr close # 会话 清理ed up

# Use named 会话s for parallel 自动化 kbr --会话 代理1 open https://site-a.com kbr --会话 代理2 open https://site-b.com

Essential Commands # Navigation kbr open # Navigate to URL kbr close # Close browser 会话 (alias: 停止) kbr re启动 # Re启动 browser 会话 kbr back # Go back in 历史 kbr forward # Go forward in 历史 kbr reload # Reload current page

# Snapshot kbr snapshot -i # Interactive elements with IDs (recommended) kbr snapshot -c # Compact mode: omit unnamed wr应用ers kbr snapshot --selector "#content" # Scope to CSS selector kbr snapshot --depth 3 # Limit tree depth kbr snapshot --ocr # Enable OCR for unnamed images kbr snapshot --cursor # Mark [cursor] on focused element

# Interaction (use numeric IDs from snapshot) kbr 命令行工具ck 1 # 命令行工具ck element kbr 命令行工具ck --new-tab 1 # 命令行工具ck and open in new tab kbr dbl命令行工具ck 1 # Double-命令行工具ck element kbr fill 2 "text" # Clear and type text kbr type 2 "text" # Type without clearing (应用end) kbr select 1 "option" # Select dropdown o