@kritchoff/agent-browser 0.9.52 → 1.0.1

package/bin/agent-browser.js

@@ -57,7 +57,8 @@ async function main() {
       default:
         // Pass through arbitrary commands to the agent daemon
         // e.g. "agent-browser open https://google.com"
-        await agent.command(...filteredArgs);
+        const result = await agent.command(...filteredArgs);
+        console.log(result);
         break;
     }
   } catch (error) {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@kritchoff/agent-browser",
-  "version": "0.9.52",
+  "version": "1.0.1",
   "description": "Headless browser automation CLI for AI agents",
   "type": "module",
   "main": "dist/index.js",
@@ -10,8 +10,6 @@
     "bin",
     "scripts",
     "skills",
-    "sdk.sh",
-    "start.sh",
     "docker-compose.sdk.yml"
   ],
   "bin": {
@@ -20,18 +18,10 @@
   "scripts": {
     "prepare": "husky",
     "version:sync": "node scripts/sync-version.js",
-    "version": "npm run version:sync && git add cli/Cargo.toml",
+    "version": "npm run version:sync",
     "build": "tsc",
-    "build:native": "npm run version:sync && cargo build --release --manifest-path cli/Cargo.toml && node scripts/copy-native.js",
-    "build:linux": "npm run version:sync && docker compose -f docker/docker-compose.yml run --rm build-linux",
-    "build:macos": "npm run version:sync && (cargo build --release --manifest-path cli/Cargo.toml --target aarch64-apple-darwin & cargo build --release --manifest-path cli/Cargo.toml --target x86_64-apple-darwin & wait) && cp cli/target/aarch64-apple-darwin/release/agent-browser bin/agent-browser-darwin-arm64 && cp cli/target/x86_64-apple-darwin/release/agent-browser bin/agent-browser-darwin-x64",
-    "build:windows": "npm run version:sync && docker compose -f docker/docker-compose.yml run --rm build-windows",
-    "build:all-platforms": "npm run version:sync && (npm run build:linux & npm run build:windows & wait) && npm run build:macos",
     "build:docker": "docker build -t agent-browser-builder -f docker/Dockerfile.build .",
-    "snapshot": "./scripts/snapshot_manager.sh",
-    "fast-reset": "./scripts/fast_reset.sh",
-    "sdk": "./sdk.sh",
-    "release": "npm run version:sync && npm run build && npm run build:all-platforms && npm publish",
+    "release": "npm run version:sync && npm run build && npm publish",
     "prepublishOnly": "npm run build",
     "start": "node dist/daemon.js",
     "dev": "tsx src/daemon.ts",
@@ -40,7 +30,6 @@
     "format:check": "prettier --check 'src/**/*.ts'",
     "test": "vitest run",
     "test:watch": "vitest",
-    "postinstall": "node scripts/postinstall.js",
     "changeset": "changeset",
     "ci:version": "changeset version && pnpm run version:sync && pnpm install --no-frozen-lockfile",
     "ci:publish": "pnpm run version:sync && pnpm run build && changeset publish"

package/README.sdk.md

@@ -1,129 +0,0 @@
-# @wootzapp/agent-browser SDK
-
-The official Node.js SDK for controlling the WootzApp Agent Browser environment. 
-
-This SDK provides a **Real Android Browser** (WootzApp) wrapped in a Docker container, controlled by a high-speed Playwright daemon. It is specifically designed for AI Agents to navigate the mobile web, bypassing bot detection, and generating LLM-friendly semantic trees (AXTree).
-
-## Features
-
-- **Real Mobile Environment**: Full Android 14 OS with Touch Events and Mobile Viewports.
-- **Zero-Config Setup**: The SDK automatically downloads and orchestrates the required Docker containers.
-- **Hyper-Speed Warm Boots**: Uses advanced VDI Volume Mounting to boot the environment in **< 5 seconds** after the first run.
-- **Fast Resets**: Cleans the browser state via Android userspace reboot in **~15 seconds**.
-- **Playwright Parity**: Control the mobile browser using standard Playwright commands (`click`, `type`, `waitForSelector`).
-- **Semantic AXTree**: Built-in `snapshot()` method generates a clean, text-based UI tree optimized for LLM reasoning.
-
----
-
-## Prerequisites
-
-1.  **Docker Engine**: Must be installed and running.
-    - *Linux Users*: Ensure your user is in the `docker` group (`sudo usermod -aG docker $USER`).
-2.  **Node.js**: v18+ is required.
-
----
-
-## Installation
-
-Install the SDK in your project:
-
-```bash
-npm install @kritchoff/agent-browser
-```
-
-*(Optional but recommended)* Install `tsx` to run TypeScript files natively:
-```bash
-npm install -D tsx
-```
-
----
-
-## Quick Start Guide
-
-Create a file named `agent.ts`:
-
-```typescript
-import { WootzAgent } from '@kritchoff/agent-browser';
-
-async function main() {
-  // 1. Initialize the controller
-  const agent = new WootzAgent();
-
-  console.log('🚀 Booting Environment...');
-  // First run: Downloads 3GB image and cold boots (~90s).
-  // Next run: Instant Hyper-Speed Warm Boot (~5s).
-  await agent.start();
-
-  console.log('🌐 Navigating to Google...');
-  await agent.navigate('https://google.com');
-
-  console.log('📸 Capturing Semantic Tree for LLM...');
-  const uiTree = await agent.snapshot();
-  console.log(uiTree); 
-
-  console.log('⌨️ Typing and Searching...');
-  await agent.type('textarea[name="q"]', 'WootzApp AI');
-  await agent.press('Enter');
-
-  console.log('🧹 Fast Reset for next task...');
-  // Wipes all tabs, cookies, and cache in ~15s
-  await agent.reset(); 
-
-  console.log('🛑 Shutting down...');
-  // Completely destroys containers and releases ports
-  await agent.stop();
-}
-
-main().catch(console.error);
-```
-
-Run your agent:
-```bash
-npx tsx agent.ts
-```
-
----
-
-## CLI Usage (Global Install)
-
-You can also use the SDK directly from your terminal to debug or control the browser manually.
-
-```bash
-npm install -g @kritchoff/agent-browser
-
-# Start the environment
-agent-browser start
-
-# Run commands
-agent-browser navigate https://news.ycombinator.com
-agent-browser click ".titleline a"
-agent-browser snapshot
-
-# Clean the browser
-agent-browser reset
-
-# Stop
-agent-browser stop
-```
-
----
-
-## Troubleshooting
-
-### `Error: Failed to connect to agent daemon (ECONNREFUSED)`
-- **Cause**: The container failed to bind port `32001` to your host machine.
-- **Fix**: Run `agent.stop()` or `docker rm -f $(docker ps -aq)` to clear old/stuck containers, then run `agent.start()` again. The SDK has built-in self-healing, but a manual hard reset always works.
-
-### `net::ERR_NAME_NOT_RESOLVED`
-- **Cause**: The Android Emulator temporarily lost its internet connection after a Warm Boot.
-- **Fix**: The SDK automatically toggles Airplane Mode to fix this, but if it persists, ensure your host machine has a stable internet connection before starting the agent.
-
-### `Selector "..." matched X elements (Strict Mode Violation)`
-- **Cause**: Playwright requires selectors to point to exactly one element.
-- **Fix**: Use more specific selectors, or use Playwright's `>> nth=0` pseudo-selector to pick the first match (e.g., `agent.click('a >> nth=0')`).
-
----
-
-## Next Steps
-
-For a complete list of all available commands (clicking, typing, tabbing, network interception), please read the [COMMANDS.md](./COMMANDS.md) file.

package/scripts/copy-native.js

@@ -1,36 +0,0 @@
-#!/usr/bin/env node
-
-/**
- * Copies the compiled Rust binary to bin/ with platform-specific naming
- */
-
-import { copyFileSync, existsSync, mkdirSync } from 'fs';
-import { dirname, join } from 'path';
-import { fileURLToPath } from 'url';
-import { platform, arch } from 'os';
-
-const __dirname = dirname(fileURLToPath(import.meta.url));
-const projectRoot = join(__dirname, '..');
-
-const sourceExt = platform() === 'win32' ? '.exe' : '';
-const sourcePath = join(projectRoot, `cli/target/release/agent-browser${sourceExt}`);
-const binDir = join(projectRoot, 'bin');
-
-// Determine platform suffix
-const platformKey = `${platform()}-${arch()}`;
-const ext = platform() === 'win32' ? '.exe' : '';
-const targetName = `agent-browser-${platformKey}${ext}`;
-const targetPath = join(binDir, targetName);
-
-if (!existsSync(sourcePath)) {
-  console.error(`Error: Native binary not found at ${sourcePath}`);
-  console.error('Run "cargo build --release --manifest-path cli/Cargo.toml" first');
-  process.exit(1);
-}
-
-if (!existsSync(binDir)) {
-  mkdirSync(binDir, { recursive: true });
-}
-
-copyFileSync(sourcePath, targetPath);
-console.log(`✓ Copied native binary to ${targetPath}`);

package/scripts/fast_reset.sh

@@ -1,117 +0,0 @@
-#!/bin/bash
-# Fast Android environment reset using userspace reboot.
-#
-# This script resets the Android emulator state much faster (~15s) than
-# a full container restart (~60s). It uses 'adb reboot userspace' to
-# restart the Android framework while keeping the kernel running.
-#
-# Usage:
-#   ./scripts/fast_reset.sh
-
-set -e
-
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-PROJECT_DIR="$(dirname "$SCRIPT_DIR")"
-
-cd "$PROJECT_DIR"
-
-# Colors for output
-RED='\033[0;31m'
-GREEN='\033[0;32m'
-YELLOW='\033[1;33m'
-BLUE='\033[0;34m'
-NC='\033[0m' # No Color
-
-log_info() {
-    echo -e "${BLUE}[INFO]${NC} $1"
-}
-
-log_success() {
-    echo -e "${GREEN}[OK]${NC} $1"
-}
-
-log_warn() {
-    echo -e "${YELLOW}[WARN]${NC} $1"
-}
-
-log_error() {
-    echo -e "${RED}[ERROR]${NC} $1"
-}
-
-# Respect COMPOSE_FILE from environment, or auto-detect
-if [ -z "$COMPOSE_FILE" ]; then
-    if [ -f "$PROJECT_DIR/docker-compose.sdk.yml" ]; then
-        COMPOSE_FILE="$PROJECT_DIR/docker-compose.sdk.yml"
-    else
-        COMPOSE_FILE="$PROJECT_DIR/docker-compose.prod.yml"
-    fi
-fi
-
-# Detect container
-CONTAINER=$(docker compose -f "$COMPOSE_FILE" ps -q android-service)
-if [ -z "$CONTAINER" ]; then
-    log_error "android-service container not running."
-    exit 1
-fi
-
-ADB_CMD="docker exec $CONTAINER adb"
-
-log_info "Initiating fast reset (userspace reboot)..."
-
-# 1. Trigger userspace reboot
-# This command returns immediately and the device goes offline
-$ADB_CMD shell reboot userspace || true
-
-# 2. Wait for device to come back online
-log_info "Waiting for device to come online..."
-start_time=$(date +%s)
-timeout=30
-
-while true; do
-    current_time=$(date +%s)
-    elapsed=$((current_time - start_time))
-    
-    if [ $elapsed -gt $timeout ]; then
-        log_error "Timeout waiting for device after ${timeout}s"
-        exit 1
-    fi
-
-    # Check if device is visible to ADB and state is 'device'
-    if $ADB_CMD get-state 2>/dev/null | grep -q "device"; then
-        # Verify shell is responsive
-        if $ADB_CMD shell echo ok 2>/dev/null | grep -q "ok"; then
-            break
-        fi
-    fi
-    
-    sleep 1
-done
-
-log_success "Device online (${elapsed}s)"
-
-# 3. Wait for CDP (Chrome DevTools Protocol)
-log_info "Waiting for browser CDP..."
-cdp_timeout=30
-cdp_start_time=$(date +%s)
-
-while true; do
-    current_time=$(date +%s)
-    elapsed=$((current_time - cdp_start_time))
-    
-    if [ $elapsed -gt $cdp_timeout ]; then
-        log_warn "Timeout waiting for CDP. Browser might not have autostarted."
-        log_info "Attempting to start browser manually..."
-        $ADB_CMD shell am start -n com.wootzapp.web/com.aspect.chromium.ChromiumMain -a android.intent.action.VIEW -d 'about:blank'
-        sleep 2
-    fi
-
-    # Check CDP version endpoint
-    if docker exec $CONTAINER curl -s --connect-timeout 2 http://localhost:9224/json/version >/dev/null; then
-        break
-    fi
-    
-    sleep 1
-done
-
-log_success "Browser CDP ready"
-log_success "Fast reset complete!"

package/scripts/postinstall.js

@@ -1,235 +0,0 @@
-#!/usr/bin/env node
-
-/**
- * Postinstall script for agent-browser
- * 
- * Downloads the platform-specific native binary if not present.
- * On global installs, patches npm's bin entry to use the native binary directly:
- * - Windows: Overwrites .cmd/.ps1 shims
- * - Mac/Linux: Replaces symlink to point to native binary
- */
-
-import { existsSync, mkdirSync, chmodSync, createWriteStream, unlinkSync, writeFileSync, symlinkSync, lstatSync } from 'fs';
-import { dirname, join } from 'path';
-import { fileURLToPath } from 'url';
-import { platform, arch } from 'os';
-import { get } from 'https';
-import { execSync } from 'child_process';
-
-const __dirname = dirname(fileURLToPath(import.meta.url));
-const projectRoot = join(__dirname, '..');
-const binDir = join(projectRoot, 'bin');
-
-// Platform detection
-const platformKey = `${platform()}-${arch()}`;
-const ext = platform() === 'win32' ? '.exe' : '';
-const binaryName = `agent-browser-${platformKey}${ext}`;
-const binaryPath = join(binDir, binaryName);
-
-// Package info
-const packageJson = JSON.parse(
-  (await import('fs')).readFileSync(join(projectRoot, 'package.json'), 'utf8')
-);
-const version = packageJson.version;
-
-// GitHub release URL
-const GITHUB_REPO = 'vercel-labs/agent-browser';
-const DOWNLOAD_URL = `https://github.com/${GITHUB_REPO}/releases/download/v${version}/${binaryName}`;
-
-async function downloadFile(url, dest) {
-  return new Promise((resolve, reject) => {
-    const file = createWriteStream(dest);
-    
-    const request = (url) => {
-      get(url, (response) => {
-        // Handle redirects
-        if (response.statusCode === 301 || response.statusCode === 302) {
-          request(response.headers.location);
-          return;
-        }
-        
-        if (response.statusCode !== 200) {
-          reject(new Error(`Failed to download: HTTP ${response.statusCode}`));
-          return;
-        }
-        
-        response.pipe(file);
-        file.on('finish', () => {
-          file.close();
-          resolve();
-        });
-      }).on('error', (err) => {
-        unlinkSync(dest);
-        reject(err);
-      });
-    };
-    
-    request(url);
-  });
-}
-
-async function main() {
-  // Check if binary already exists
-  if (existsSync(binaryPath)) {
-    // Ensure binary is executable (npm doesn't preserve execute bit)
-    if (platform() !== 'win32') {
-      chmodSync(binaryPath, 0o755);
-    }
-    console.log(`✓ Native binary ready: ${binaryName}`);
-    
-    // On global installs, fix npm's bin entry to use native binary directly
-    await fixGlobalInstallBin();
-    
-    showPlaywrightReminder();
-    return;
-  }
-
-  // Ensure bin directory exists
-  if (!existsSync(binDir)) {
-    mkdirSync(binDir, { recursive: true });
-  }
-
-  console.log(`Downloading native binary for ${platformKey}...`);
-  console.log(`URL: ${DOWNLOAD_URL}`);
-
-  try {
-    await downloadFile(DOWNLOAD_URL, binaryPath);
-    
-    // Make executable on Unix
-    if (platform() !== 'win32') {
-      chmodSync(binaryPath, 0o755);
-    }
-    
-    console.log(`✓ Downloaded native binary: ${binaryName}`);
-  } catch (err) {
-    console.log(`⚠ Could not download native binary: ${err.message}`);
-    console.log(`  The CLI will use Node.js fallback (slightly slower startup)`);
-    console.log('');
-    console.log('To build the native binary locally:');
-    console.log('  1. Install Rust: https://rustup.rs');
-    console.log('  2. Run: npm run build:native');
-  }
-
-  // On global installs, fix npm's bin entry to use native binary directly
-  // This avoids the /bin/sh error on Windows and provides zero-overhead execution
-  await fixGlobalInstallBin();
-
-  showPlaywrightReminder();
-}
-
-function showPlaywrightReminder() {
-  console.log('');
-  console.log('╔═══════════════════════════════════════════════════════════════════════════╗');
-  console.log('║ To download browser binaries, run:                                        ║');
-  console.log('║                                                                           ║');
-  console.log('║     npx playwright install chromium                                       ║');
-  console.log('║                                                                           ║');
-  console.log('║ On Linux, include system dependencies with:                               ║');
-  console.log('║                                                                           ║');
-  console.log('║     npx playwright install --with-deps chromium                           ║');
-  console.log('║                                                                           ║');
-  console.log('╚═══════════════════════════════════════════════════════════════════════════╝');
-}
-
-/**
- * Fix npm's bin entry on global installs to use the native binary directly.
- * This provides zero-overhead CLI execution for global installs.
- */
-async function fixGlobalInstallBin() {
-  if (platform() === 'win32') {
-    await fixWindowsShims();
-  } else {
-    await fixUnixSymlink();
-  }
-}
-
-/**
- * Fix npm symlink on Mac/Linux global installs.
- * Replace the symlink to the JS wrapper with a symlink to the native binary.
- */
-async function fixUnixSymlink() {
-  // Get npm's global bin directory (npm prefix -g + /bin)
-  let npmBinDir;
-  try {
-    const prefix = execSync('npm prefix -g', { encoding: 'utf8' }).trim();
-    npmBinDir = join(prefix, 'bin');
-  } catch {
-    return; // npm not available
-  }
-
-  const symlinkPath = join(npmBinDir, 'agent-browser');
-
-  // Check if symlink exists (indicates global install)
-  try {
-    const stat = lstatSync(symlinkPath);
-    if (!stat.isSymbolicLink()) {
-      return; // Not a symlink, don't touch it
-    }
-  } catch {
-    return; // Symlink doesn't exist, not a global install
-  }
-
-  // Replace symlink to point directly to native binary
-  try {
-    unlinkSync(symlinkPath);
-    symlinkSync(binaryPath, symlinkPath);
-    console.log('✓ Optimized: symlink points to native binary (zero overhead)');
-  } catch (err) {
-    // Permission error or other issue - not critical, JS wrapper still works
-    console.log(`⚠ Could not optimize symlink: ${err.message}`);
-    console.log('  CLI will work via Node.js wrapper (slightly slower startup)');
-  }
-}
-
-/**
- * Fix npm-generated shims on Windows global installs.
- * npm generates shims that try to run /bin/sh, which doesn't exist on Windows.
- * We overwrite them to invoke the native .exe directly.
- */
-async function fixWindowsShims() {
-  // Check if this is a global install by looking for npm's global prefix
-  let npmBinDir;
-  try {
-    npmBinDir = execSync('npm prefix -g', { encoding: 'utf8' }).trim();
-  } catch {
-    return; // Not a global install or npm not available
-  }
-
-  // The shims are in the npm prefix directory (not prefix/bin on Windows)
-  const cmdShim = join(npmBinDir, 'agent-browser.cmd');
-  const ps1Shim = join(npmBinDir, 'agent-browser.ps1');
-
-  // Only fix if shims exist (indicates global install)
-  if (!existsSync(cmdShim)) {
-    return;
-  }
-
-  // Path to native binary relative to npm prefix
-  const relativeBinaryPath = 'node_modules\\agent-browser\\bin\\agent-browser-win32-x64.exe';
-
-  try {
-    // Overwrite .cmd shim
-    const cmdContent = `@ECHO off\r\n"%~dp0${relativeBinaryPath}" %*\r\n`;
-    writeFileSync(cmdShim, cmdContent);
-
-    // Overwrite .ps1 shim
-    const ps1Content = `#!/usr/bin/env pwsh
-$basedir = Split-Path $MyInvocation.MyCommand.Definition -Parent
-$exe = ""
-if ($PSVersionTable.PSVersion -lt "6.0" -or $IsWindows) {
-  $exe = ".exe"
-}
-& "$basedir/${relativeBinaryPath.replace(/\\/g, '/')}" $args
-exit $LASTEXITCODE
-`;
-    writeFileSync(ps1Shim, ps1Content);
-
-    console.log('✓ Optimized: shims point to native binary (zero overhead)');
-  } catch (err) {
-    // Permission error or other issue - not critical, JS wrapper still works
-    console.log(`⚠ Could not optimize shims: ${err.message}`);
-    console.log('  CLI will work via Node.js wrapper (slightly slower startup)');
-  }
-}
-
-main().catch(console.error);