원클릭으로
playwright-browser
Headless browser automation using Playwright CLI. Use for browser testing, screenshots, scraping, and parallel browser sessions. Token-efficient CLI — no MCP overhead.
Headless browser automation using Playwright CLI. Use for browser testing, screenshots, scraping, and parallel browser sessions. Token-efficient CLI — no MCP overhead.
Complete task lifecycle management system with state-based workflow. Define, refine, implement, and review tasks with automated quality gates and intelligent discovery.
Analyzes project structure, dependencies, and patterns to generate comprehensive context documentation. Uses parallel agent execution for faster analysis. Helps understand unfamiliar codebases quickly.
Comprehensive code quality verification workflow that checks linting, formatting, type safety, tests, and build before committing code. Uses parallel agent execution for maximum speed. Non-destructive - only reports issues without making changes.
Auto-activating documentation cache with fresh API docs. Fetches and automatically consults cached documentation when user works with libraries/frameworks.
Fork a terminal session to a new terminal window. Use this when the user requests 'fork terminal' or 'create a new terminal' or 'new terminal: <command>' or 'fork session: <command>'.
| name | Playwright Browser |
| description | Headless browser automation using Playwright CLI. Use for browser testing, screenshots, scraping, and parallel browser sessions. Token-efficient CLI — no MCP overhead. |
| trigger | auto |
| allowed-tools | ["Bash","Read","Write"] |
Automate browsers using playwright-cli, a token-efficient CLI for Playwright. Runs headless by default, supports parallel sessions via named sessions (-s=), and persistent profiles for cookies/state. Use when you need headless browsing, UI testing, screenshots, web scraping, or any browser automation that can run in the background.
HEADED: false # Show browser window. Options: true, false VISION: false # Return screenshots as images in context (higher token cost) VIEWPORT_SIZE: 1440x900 # Browser viewport dimensions (WxH) SCREENSHOTS_DIR: ./screenshots # Where screenshots are saved
Open Session
--persistent to preserve cookies/localStorage across commands-s=mystore-checkout-s=competitor-pricing--headed flagPLAYWRIGHT_MCP_CAPS=visionPLAYWRIGHT_MCP_VIEWPORT_SIZE=<VIEWPORT_SIZE> playwright-cli -s=<session-name> open <url> --persistent
PLAYWRIGHT_MCP_VIEWPORT_SIZE=1440x900 playwright-cli -s=mystore-checkout open https://mystore.com --persistent --headed
Snapshot Page
e12, e45) you can click, fill, etc.playwright-cli -s=mystore-checkout snapshot → returns element tree with refsplaywright-cli -s=<session-name> snapshotInteract with Elements
playwright-cli -s=mystore-checkout click e12 → clicks element ref e12playwright-cli -s=mystore-checkout fill e15 "test@example.com" → fills inputplaywright-cli -s=mystore-checkout type "search query" → types into focused elementplaywright-cli -s=mystore-checkout press Enter → presses keyboard keygoto <url>, go-back, go-forward, reloadplaywright-cli -s=<session-name> <command> <args>Take Screenshots
--filename=<path>playwright-cli -s=mystore-checkout screenshot → saves to default locationplaywright-cli -s=mystore-checkout screenshot --filename=./screenshots/checkout-step1.pngplaywright-cli -s=<session-name> screenshot [--filename=<path>]Close Session
playwright-cli -s=mystore-checkout closeplaywright-cli close-allplaywright-cli -s=<name> delete-dataplaywright-cli -s=<session-name> close| Command | Description | Example |
|---|---|---|
open [url] | Launch browser | -s=test open https://example.com --persistent |
close | Close session | -s=test close |
goto <url> | Navigate to URL | -s=test goto https://example.com/page2 |
snapshot | Get element refs | -s=test snapshot |
screenshot | Capture page image | -s=test screenshot --filename=out.png |
click <ref> | Click element | -s=test click e12 |
fill <ref> <text> | Fill input field | -s=test fill e15 "hello" |
type <text> | Type into focused | -s=test type "search" |
press <key> | Press keyboard key | -s=test press Enter |
hover <ref> | Hover over element | -s=test hover e20 |
select <ref> <val> | Select dropdown | -s=test select e30 "option1" |
go-back | Back | -s=test go-back |
go-forward | Forward | -s=test go-forward |
reload | Reload page | -s=test reload |
tab-list | List tabs | -s=test tab-list |
tab-new [url] | New tab | -s=test tab-new https://example.com |
tab-select <i> | Switch tab | -s=test tab-select 1 |
list | List sessions | list |
close-all | Close all sessions | close-all |
Common flags: --persistent (cookies/state), --headed (visible browser), --browser=chrome, --filename=<path>