agent-browser-priv 0.27.3-priv.3 → 0.27.3-priv.4

package/README.md

@@ -1,7 +1,8 @@
-# agent-browser-priv
+# agent-browser
 
-Browser automation CLI for AI agents, forked for opt-in local privacy/runtime
-backends. Fast native Rust CLI.
+Browser automation CLI for AI agents, distributed from the
+`agent-browser-priv` fork with Patchright as the default local backend. Fast
+native Rust CLI.
 
 [![skills.sh](https://skills.sh/b/vercel-labs/agent-browser)](https://skills.sh/vercel-labs/agent-browser)
 
@@ -13,7 +14,7 @@ Installs the native Rust binary:
 
 ```bash
 npm install -g agent-browser-priv
-agent-browser-priv install  # Download Chrome from Chrome for Testing (first time only)
+agent-browser install  # Install Patchright browser artifacts (first time only)
 ```
 
 ### Project Installation (local dependency)
@@ -22,7 +23,7 @@ For projects that want to pin the version in `package.json`:
 
 ```bash
 npm install agent-browser-priv
-agent-browser-priv install
+agent-browser install
 ```
 
 Then use via `package.json` scripts or by invoking `agent-browser` directly.
@@ -30,15 +31,15 @@ Then use via `package.json` scripts or by invoking `agent-browser` directly.
 ### Homebrew (macOS)
 
 ```bash
-brew install liuwen/agent-browser-priv/agent-browser-priv
-agent-browser-priv install  # Download Chrome from Chrome for Testing (first time only)
+brew install liuwen/agent-browser-priv/agent-browser
+agent-browser install  # Install Patchright browser artifacts (first time only)
 ```
 
 ### Cargo (Rust)
 
 ```bash
 cargo install --git https://github.com/liuwen/agent-browser-priv
-agent-browser-priv install  # Download Chrome from Chrome for Testing (first time only)
+agent-browser install  # Install Patchright browser artifacts (first time only)
 ```
 
 ### From Source
@@ -51,8 +52,8 @@ cd agent-browser-priv
 pnpm install
 pnpm build
 pnpm build:native   # Requires Rust (https://rustup.rs)
-pnpm link --global  # Makes agent-browser-priv available globally
-agent-browser-priv install
+pnpm link --global  # Makes agent-browser available globally
+agent-browser install
 ```
 
 ### Linux Dependencies
@@ -60,7 +61,7 @@ agent-browser-priv install
 On Linux, install system dependencies:
 
 ```bash
-agent-browser-priv install --with-deps
+agent-browser install --with-deps
 ```
 
 ### Updating
@@ -68,14 +69,15 @@ agent-browser-priv install --with-deps
 Upgrade to the latest version:
 
 ```bash
-agent-browser-priv upgrade
+agent-browser upgrade
 ```
 
 Detects your installation method (npm, Homebrew, or Cargo) and runs the appropriate update command automatically.
 
 ### Requirements
 
-- **Chrome** - Run `agent-browser-priv install` to download Chrome from [Chrome for Testing](https://developer.chrome.com/blog/chrome-for-testing/) (Google's official automation channel). Existing Chrome, Brave, Playwright, and Puppeteer installations are detected automatically. No Playwright or Node.js required for the daemon.
+- **Patchright backend** - Run `agent-browser install` to install Patchright and its browser artifacts for the default local backend.
+- **Chrome backend** - Use `agent-browser install chrome` to download Chrome from [Chrome for Testing](https://developer.chrome.com/blog/chrome-for-testing/) for `--backend chrome`. Existing Chrome, Brave, Playwright, and Puppeteer installations are detected automatically.
 - **Node.js 24+ and pnpm 11+** - Only needed when building from source.
 - **Rust** - Only needed when building from source (see From Source above).
 
@@ -458,18 +460,19 @@ agent-browser removeinitscript <identifier>       # Remove a previously register
 ### Setup
 
 ```bash
-agent-browser install                 # Download Chrome from Chrome for Testing (Google's official automation channel)
+agent-browser install                 # Install Patchright browser artifacts
+agent-browser install chrome          # Download Chrome from Chrome for Testing
 agent-browser install --with-deps     # Also install system deps (Linux)
 agent-browser upgrade                 # Upgrade agent-browser to the latest version
 agent-browser doctor                  # Diagnose the install and auto-clean stale daemon files
-agent-browser doctor --fix            # Also run destructive repairs (reinstall Chrome, purge old state, ...)
+agent-browser doctor --fix            # Also run destructive repairs
 agent-browser doctor --offline --quick  # Skip network probes and the live launch test
 ```
 
-`doctor` checks your environment, Chrome install, daemon state, config files,
-encryption key, providers, network reachability, and runs a live headless
-browser launch test. Stale socket/pid sidecar files are auto-cleaned. Output
-is also available as `--json` for agents.
+`doctor` checks your environment, browser backend install, daemon state,
+config files, encryption key, providers, network reachability, and runs a live
+headless browser launch test. Stale socket/pid sidecar files are auto-cleaned.
+Output is also available as `--json` for agents.
 
 ### Skills
 
@@ -743,7 +746,7 @@ This is useful for multimodal AI models that can reason about visual layout, unl
 | `--confirm-actions <list>` | Action categories requiring confirmation (or `AGENT_BROWSER_CONFIRM_ACTIONS` env) |
 | `--confirm-interactive` | Interactive confirmation prompts; auto-denies if stdin is not a TTY (or `AGENT_BROWSER_CONFIRM_INTERACTIVE` env) |
 | `--engine <name>` | Browser engine: `chrome` (default), `lightpanda` (or `AGENT_BROWSER_ENGINE` env) |
-| `--backend <name>` | Local Chrome backend: `chrome` (default), `patchright` (or `AGENT_BROWSER_BACKEND` env) |
+| `--backend <name>` | Local Chrome backend: `patchright` (default), `chrome` (or `AGENT_BROWSER_BACKEND` env) |
 | `--no-auto-dialog` | Disable automatic dismissal of `alert`/`beforeunload` dialogs (or `AGENT_BROWSER_NO_AUTO_DIALOG` env) |
 | `--model <name>` | AI model for chat command (or `AI_GATEWAY_MODEL` env) |
 | `-v`, `--verbose` | Show tool commands and their raw output (chat) |
@@ -751,27 +754,33 @@ This is useful for multimodal AI models that can reason about visual layout, unl
 | `--config <path>` | Use a custom config file (or `AGENT_BROWSER_CONFIG` env) |
 | `--debug` | Debug output |
 
-## Patchright backend
+## Local backends
 
-`agent-browser-priv` keeps the normal Chrome CDP backend by default. For local
-development environments that need Patchright-managed Chromium artifacts or a
-Patchright-launched persistent browser, install the backend once:
+This fork defaults local Chrome-compatible launches to Patchright. For fresh
+remote hosts, sandboxes, and CI environments, install the default backend once:
 
 ```bash
-agent-browser-priv install patchright
+agent-browser install
 ```
 
-Then opt in per session:
+Then use the normal command surface:
 
 ```bash
-agent-browser-priv --backend patchright --headed open https://example.com
-agent-browser-priv --backend patchright --profile ~/.agent-browser-priv/profiles/dev open https://example.com
+agent-browser --headed open https://example.com
+agent-browser --profile ~/.agent-browser/profiles/dev open https://example.com
+```
+
+Use the built-in Chrome CDP launcher when a site behaves better on that lane:
+
+```bash
+agent-browser install chrome
+agent-browser --backend chrome open https://example.com
+AGENT_BROWSER_BACKEND=chrome agent-browser open https://example.com
 ```
 
 Patchright is used only to launch the local Chrome-compatible browser and expose
-CDP on localhost. The agent-browser command surface remains unchanged. This
-backend does not solve CAPTCHA, Turnstile, or other human verification pages;
-preserve those pages for human handoff.
+CDP on localhost. It does not solve CAPTCHA, Turnstile, or other human
+verification pages; preserve those pages for human handoff.
 
 ## Observability Dashboard
 
@@ -1255,7 +1264,7 @@ agent-browser uses a client-daemon architecture:
 
 The daemon starts automatically on first command and persists between commands for fast subsequent operations. To auto-shutdown the daemon after a period of inactivity, set `AGENT_BROWSER_IDLE_TIMEOUT_MS` (value in milliseconds). When set, the daemon closes the browser and exits after receiving no commands for the specified duration.
 
-**Browser Engine:** Uses Chrome (from Chrome for Testing) by default. The `--engine` flag selects between `chrome` and `lightpanda`. Supported browsers: Chromium/Chrome (via CDP) and Safari (via WebDriver for iOS).
+**Browser Engine:** Uses the Chrome-compatible engine with Patchright by default. The `--engine` flag selects between `chrome` and `lightpanda`; the `--backend` flag selects `patchright` or `chrome` for local Chrome-compatible launches. Supported browsers: Chromium/Chrome (via CDP) and Safari (via WebDriver for iOS).
 
 ## Platforms
 

package/bin/agent-browser.js

@@ -1,7 +1,7 @@
 #!/usr/bin/env node
 
 /**
- * Cross-platform CLI wrapper for agent-browser-priv
+ * Cross-platform CLI wrapper for the agent-browser-priv npm package.
  * 
  * This wrapper enables npx support on Windows where shell scripts don't work.
  * For global installs, postinstall.js patches the shims to invoke the native

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "agent-browser-priv",
-  "version": "0.27.3-priv.3",
+  "version": "0.27.3-priv.4",
   "description": "Browser automation CLI for AI agents",
   "type": "module",
   "packageManager": "pnpm@11.1.3",
@@ -15,6 +15,7 @@
     "skills"
   ],
   "bin": {
+    "agent-browser": "bin/agent-browser.js",
     "agent-browser-priv": "bin/agent-browser.js"
   },
   "scripts": {

package/scripts/postinstall.js

@@ -1,7 +1,7 @@
 #!/usr/bin/env node
 
 /**
- * Postinstall script for agent-browser-priv
+ * Postinstall script for the agent-browser-priv npm package.
  * 
  * Downloads the platform-specific native binary if not present.
  * On global installs, patches npm's bin entry to use the native binary directly:
@@ -91,7 +91,7 @@ async function downloadFile(url, dest) {
 
 /**
  * Detect which package manager ran this postinstall and write a marker file
- * next to the binary so `agent-browser-priv upgrade` can use the correct one
+ * next to the binary so `agent-browser upgrade` can use the correct one
  * without fragile path heuristics or slow subprocess probing.
  *
  * npm_config_user_agent is set by npm/pnpm/yarn/bun during lifecycle scripts,
@@ -205,21 +205,25 @@ function showInstallReminder() {
   if (systemChrome) {
     console.log('');
     console.log(`  ✓ System Chrome found: ${systemChrome}`);
-    console.log('    agent-browser-priv will use it automatically.');
+    console.log('    Use --backend chrome or config backend=chrome to launch it directly.');
+    console.log('');
+    console.log('  The default Patchright backend is prepared with:');
+    console.log('');
+    console.log('    agent-browser install');
     console.log('');
     return;
   }
 
   console.log('');
-  console.log('  ⚠ No Chrome installation detected.');
-  console.log('  If you plan to use a local browser, run:');
+  console.log('  ⚠ No system Chrome installation detected.');
+  console.log('  To prepare the default Patchright backend, run:');
   console.log('');
-  console.log('    agent-browser-priv install');
+  console.log('    agent-browser install');
   if (platform() === 'linux') {
     console.log('');
     console.log('  On Linux, include system dependencies with:');
     console.log('');
-    console.log('    agent-browser-priv install --with-deps');
+    console.log('    agent-browser install --with-deps');
   }
   console.log('');
   console.log('  You can skip this if you use --cdp, --provider, --engine, or --executable-path.');
@@ -252,27 +256,29 @@ async function fixUnixSymlink() {
     return; // npm not available
   }
 
-  const symlinkPath = join(npmBinDir, 'agent-browser-priv');
+  const symlinkPaths = [
+    join(npmBinDir, 'agent-browser'),
+    join(npmBinDir, 'agent-browser-priv'),
+  ];
 
-  // Check if symlink exists (indicates global install)
-  try {
-    const stat = lstatSync(symlinkPath);
-    if (!stat.isSymbolicLink()) {
-      return; // Not a symlink, don't touch it
+  let optimized = false;
+  for (const symlinkPath of symlinkPaths) {
+    try {
+      const stat = lstatSync(symlinkPath);
+      if (!stat.isSymbolicLink()) {
+        continue;
+      }
+
+      unlinkSync(symlinkPath);
+      symlinkSync(binaryPath, symlinkPath);
+      optimized = true;
+    } catch {
+      // Symlink missing during postinstall or not writable; JS wrapper still works.
     }
-  } 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)');
+  if (optimized) {
+    console.log('✓ Optimized: command symlink points to native binary (zero overhead)');
   }
 }
 
@@ -289,16 +295,6 @@ async function fixWindowsShims() {
     return;
   }
 
-  const cmdShim = join(npmBinDir, 'agent-browser-priv.cmd');
-  const ps1Shim = join(npmBinDir, 'agent-browser-priv.ps1');
-
-  // Shims may not exist yet during postinstall (npm creates them after
-  // lifecycle scripts). If missing, fall back: the JS wrapper at
-  // bin/agent-browser.js handles Windows correctly via child_process.spawn.
-  if (!existsSync(cmdShim)) {
-    return;
-  }
-
   // Detect architecture so ARM64 Windows is handled correctly
   // (falls back to x64 binary — see platform detection above)
   const cpuArch = effectiveArch;
@@ -310,17 +306,24 @@ async function fixWindowsShims() {
     return;
   }
 
-  try {
+  let optimized = false;
+  for (const commandName of ['agent-browser', 'agent-browser-priv']) {
+    const cmdShim = join(npmBinDir, `${commandName}.cmd`);
+    const ps1Shim = join(npmBinDir, `${commandName}.ps1`);
+    if (!existsSync(cmdShim)) {
+      continue;
+    }
+
     const cmdContent = `@ECHO off\r\n"%~dp0${relativeBinaryPath}" %*\r\n`;
     writeFileSync(cmdShim, cmdContent);
 
     const ps1Content = `#!/usr/bin/env pwsh\r\n$basedir = Split-Path $MyInvocation.MyCommand.Definition -Parent\r\n& "$basedir\\${relativeBinaryPath}" $args\r\nexit $LASTEXITCODE\r\n`;
     writeFileSync(ps1Shim, ps1Content);
+    optimized = true;
+  }
 
+  if (optimized) {
     console.log('✓ Optimized: shims point to native binary (zero overhead)');
-  } catch (err) {
-    console.log(`⚠ Could not optimize shims: ${err.message}`);
-    console.log('  CLI will work via Node.js wrapper (slightly slower startup)');
   }
 }
 

package/skill-data/core/SKILL.md

@@ -1,20 +1,19 @@
 ---
 name: core
 description: Core agent-browser usage guide. Read this before running any agent-browser commands. Covers the snapshot-and-ref workflow, navigating pages, interacting with elements (click, fill, type, select), extracting text and data, taking screenshots, managing tabs, handling forms and auth, waiting for content, running multiple browser sessions in parallel, and troubleshooting common failures. Use when the user asks to interact with a website, fill a form, click something, extract data, take a screenshot, log into a site, test a web app, or automate any browser task.
-allowed-tools: Bash(agent-browser:*), Bash(npx agent-browser:*), Bash(agent-browser-priv:*)
+allowed-tools: Bash(agent-browser:*), Bash(agent-browser-priv:*), Bash(npx agent-browser-priv:*)
 ---
 
 # agent-browser core
 
-Fast browser automation CLI for AI agents. Chrome/Chromium via CDP, no
-Playwright or Puppeteer dependency. Accessibility-tree snapshots with compact
-`@eN` refs let agents interact with pages in ~200-400 tokens instead of
-parsing raw HTML.
+Fast browser automation CLI for AI agents. This fork is distributed as the
+`agent-browser-priv` npm package, but it intentionally installs the
+`agent-browser` command for daily use. Local Chrome-compatible launches use
+Patchright by default; use `--backend chrome` for the built-in Chrome CDP
+launcher when that lane behaves better for a target. Accessibility-tree
+snapshots with compact `@eN` refs let agents interact with pages in ~200-400
+tokens instead of parsing raw HTML.
 
-When `agent-browser-priv` is available, use it only for explicitly requested
-privacy/local-runtime work or authorized local development debugging where the
-normal browser lane hits local-only bot/challenge friction. The command surface
-is the same; opt into Patchright with `agent-browser-priv --backend patchright`.
 Do not add CAPTCHA solving, Turnstile solving, proxy rotation policy, or
 production stealth defaults. If a challenge remains, classify it and preserve
 artifacts for human handoff.
@@ -40,8 +39,8 @@ next ref interaction.
 ## Quickstart
 
 ```bash
-# Install once
-npm i -g agent-browser && agent-browser install
+# Install once from this fork
+npm i -g agent-browser-priv && agent-browser install
 
 # Take a screenshot of a page
 agent-browser open https://example.com
@@ -343,13 +342,13 @@ agent-browser dialog dismiss          # cancel
 ## Diagnosing install issues
 
 If a command fails unexpectedly (`Unknown command`, `Failed to connect`,
-stale daemons, version mismatches after `upgrade`, missing Chrome, etc.)
+stale daemons, version mismatches after `upgrade`, missing browser backend, etc.)
 run `doctor` before anything else:
 
 ```bash
-agent-browser doctor                     # full diagnosis (env, Chrome, daemons, config, providers, network, launch test)
+agent-browser doctor                     # full diagnosis (env, browser backend, daemons, config, providers, network, launch test)
 agent-browser doctor --offline --quick   # fast, local-only
-agent-browser doctor --fix               # also run destructive repairs (reinstall Chrome, purge old state, ...)
+agent-browser doctor --fix               # also run destructive repairs (install browser backends, purge old state, ...)
 agent-browser doctor --json              # structured output for programmatic consumption
 ```
 
@@ -419,7 +418,7 @@ and [references/authentication.md](references/authentication.md).
 --headed                # show the window (default is headless)
 --auto-connect          # connect to an already-running Chrome
 --cdp <port>            # connect to a specific CDP port
---backend <name>        # agent-browser-priv local backend: chrome, patchright
+--backend <name>        # local backend: patchright (default), chrome
 --profile <name|path>   # use a Chrome profile (login state survives)
 --headers <json>        # HTTP headers scoped to the URL's origin
 --proxy <url>           # proxy server