browserforce

Original：🇺🇸 English
Translated

Browse the web using the user's real Chrome browser — already logged in, with real cookies and extensions. No headless browser. Uses BrowserForce relay + Playwright API via CLI.

2installs
Sourceivalsaraj/browserforce
Added on

NPX Install

npx skill4agent add ivalsaraj/browserforce browserforce

Tags

Translated version includes tags in frontmatter
browserforceplaywright-apicli-toolreal-browser-automationweb-automation

SKILL.md Content

View Translation Comparison →

BrowserForce — Your Real Chrome Browser

BrowserForce gives you the user's actual Chrome browser — all their logins, cookies, and extensions already active. No headless browser, no fresh profiles.

Prerequisites

The user must have:
  1. BrowserForce Chrome extension installed and connected (green icon)
  2. The relay auto-starts on first command — no manual step needed
Check with: 
browserforce status

Quick Reference

bash
browserforce status              # Check relay + extension status
browserforce tabs                # List open tabs
browserforce snapshot [n]        # Accessibility tree of tab n
browserforce screenshot [n]      # Screenshot tab n (PNG to stdout)
browserforce navigate <url>      # Open URL in new tab
browserforce -e "<code>"         # Run Playwright JavaScript (one-shot)

Important: One-Shot Execution

Each 
browserforce -e
call is independent — state does NOT persist between calls. Do everything you need (navigate, act, observe) within a single 
-e
call.

Core Workflow: Observe → Act → Observe

Quick observation (no code needed)

bash
browserforce snapshot 0          # See what's on tab 0
browserforce tabs                # List all tabs

Navigate and read a page

bash
browserforce -e "
  state.page = await context.newPage();
  await state.page.goto('https://example.com');
  await waitForPageLoad();
  return await snapshot();
"
Note: 
snapshot()
reads from 
state.page
(if set) or 
page
(default tab 0). Always assign 
state.page
when creating a new page so 
snapshot()
reads the right tab.

Click and verify

bash
browserforce -e "
  state.page = context.pages()[context.pages().length - 1];
  await state.page.locator('role=button[name=\"Next\"]').click();
  await waitForPageLoad();
  return await snapshot();
"

Fill a form

bash
browserforce -e "
  state.page = context.pages()[context.pages().length - 1];
  await state.page.locator('role=textbox[name=\"Email\"]').fill('user@example.com');
  return await snapshot();
"

Extract data

bash
browserforce -e "
  const p = context.pages()[context.pages().length - 1];
  return await p.evaluate(() => document.querySelector('.price').textContent);
"

Screenshot

bash
browserforce screenshot 0 > page.png
# or via -e:
browserforce -e "
  state.page = context.pages()[0];
  return await state.page.screenshot();
" > page.png

Rules

  1. snapshot() over screenshot() — snapshot returns text (fast, cheap). Use screenshot only for visual layout verification.
  2. One-shot execution — each -e call is independent. Do all steps in one call.
  3. Don't navigate existing tabs — create your own via 
    context.newPage()
    or 
    browserforce navigate <url>
    .
  4. Use convenience commands for simple operations: 
    browserforce tabs
    , 
    browserforce snapshot
    , 
    browserforce screenshot
    .
  5. waitForPageLoad() — call after navigation or clicks that trigger page loads.

Error Recovery

  • Connection lost: User must check 
    browserforce status
  • No tabs: 
    browserforce navigate https://example.com
  • Element not found: 
    browserforce -e "return await snapshot({ search: 'button' })"

Important

This is the user's REAL browser. Be respectful:
  • Don't close tabs you didn't open
  • Don't navigate tabs you didn't create
  • Don't modify browser settings or stored data