Playwright Agents
Introducing Playwright Agents, three custom agent definitions designed to guide LLMs through the core process of building a Playwright test:
- 🎭 planner explores the app and produces a Markdown test plan
- 🎭 generator transforms the Markdown plan into the Playwright Test files
- 🎭 healer executes the test suite and automatically repairs failing tests
Run npx playwright init-agents
with your client of choice to generate the latest agent definitions:
# Generate agent files for each agentic loop
# Visual Studio Code
npx playwright init-agents --loop=vscode
# Claude Code
npx playwright init-agents --loop=claude
# opencode
npx playwright init-agents --loop=opencode
Learn more about Playwright Agents
New APIs
- New methods page.consoleMessages() and page.pageErrors() for retrieving the most recent console messages from the page
- New method page.requests() for retrieving the most recent network requests from the page
- Added
--test-list
and--test-list-invert
to allow manual specification of specific tests from a file
UI Mode and HTML Reporter
- Added option to
'html'
reporter to disable the "Copy prompt" button - Added option to
'html'
reporter and UI Mode to merge files, collapsing test and describe blocks into a single unified list - Added option to UI Mode mirroring the
--update-snapshots
options - Added option to UI Mode to run only a single worker at a time
Breaking Changes
- Event browserContext.on('backgroundpage') has been deprecated and will not be emitted. Method browserContext.backgroundPages() will return an empty list
Miscellaneous
- Aria snapshots render and compare
input
placeholder
- Added environment variable
PLAYWRIGHT_TEST
to Playwright worker processes to allow discriminating on testing status
Browser Versions
- Chromium 141.0.7390.37
- Mozilla Firefox 142.0.1
- WebKit 26.0