microsoft playwright
OFFICIALLABSCO SUMMARY
playwright-cli automates browser interactions and runs Playwright tests, and works for anyone using Playwright. playwright-trace reads a Playwright trace file from the command line — actions, requests, console output, errors, snapshots — which is useful to any test author debugging a failure, not just the Playwright team.
The other four are the Playwright team's own maintainer tooling. playwright-dev explains how to extend Playwright itself — adding APIs, MCP tools, or CLI commands. playwright-devops handles CI failure analysis, workflow debugging, and releases for the microsoft/playwright repo specifically. playwright-triage reproduces a bug from a GitHub issue's description. playwright-bot-voice is the house style guide for anything the Playwright bot posts publicly under its own name — issue comments, PR descriptions, release notes — so agent-written text reads as one maintainer's voice rather than a model's.
READ THE FULL ANALYSIS
This is a much narrower slice than the Playwright README describes. The README covers Playwright Test, the Playwright CLI package, Playwright MCP, the Playwright Library, and a VS Code extension — five separate ways to use Playwright. The six skills in this repo don't teach any of that directly; they are a bug-triage and repo-maintenance layer sitting on top, written for people who work on Playwright's own codebase.
playwright-devops needs GitHub access it doesn't advertise up front. Downloading failed CI job logs requires reachable GitHub Actions credentials — fine inside Microsoft's own CI, not something an outside reader can just run.
Checked 19 September 2026 against the live skill set and the repository README.
ALSO IN THIS PACKAGE
playwright test coverage.prompt
Produce test coverage
playwright test generate.prompt
Generate test plan
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright
playwright test heal.prompt
Fix tests
playwright-test-healer
Use this agent when you need to debug and fix failing Playwright tests
playwright test plan.prompt
Create test plan
playwright-test-planner
Use this agent when you need to create comprehensive test plan for a web application or website
WHAT'S INSIDE
5 showing · 5 totalNothing else to set up — install it and go.
playwright-cli
Instead of writing a script to make a browser do something, you type the command and the browser does it right away — one step at a time, on a real live page.
playwright-dev
Where do you go to change Playwright itself? This names the corner of the source code each kind of change lives in, and how to build and test it once it is written.
playwright-devops
Finding out why Playwright's own build failed on the latest change to the project, by pulling down the logs of the jobs that broke and reading them.
playwright-trace
A failed test leaves behind a recording of everything the browser did, and this reads it back without opening a browser: every action, every network call, every error message.
playwright-triage
Takes a bug someone reported against Playwright and settles what it really is — reproduced, already fixed, not reproducible, or working as intended — and stops short of fixing it.
HOW TO GET IT
npx skills add microsoft/playwrightnpx skills add microsoft/playwright --skill <name> --full-depthPick the skill name from the Skills tab — each entry there installs independently.