Playwright

Skill mblauberg/provenant/skills/playwright

Portable equal-primary Claude and Codex agent harness for a governed agentic SDLC

Install

npx -y skills add mblauberg/provenant --skill playwright

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

27 days oldThe repository was created 27 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use for terminal-driven browser navigation, interaction, snapshots, screenshots, extraction, or UI-flow debugging. Not for authoring Playwright test suites or frontend design; use tdd or ui-ux-design.

SKILL.md

2.9 KB, as published. Nobody here has run it

Playwright CLI Skill

Drive a real browser from the terminal with playwright-cli and the bundled wrapper. Do not pivot to @playwright/test unless explicitly asked for tests.

Prerequisite check (required)

Prefer an existing playwright-cli. The wrapper fails closed rather than installing from the network unless the user authorises package retrieval:

command -v playwright-cli || command -v npx

If both are absent, report the missing capability. Do not install Node or a CLI without authority. For an authorised one-run npx resolution, set:

PLAYWRIGHT_CLI_ALLOW_NPX_INSTALL=1 "$PWCLI" --help

An existing PATH binary is used as installed; the wrapper does not attest its version. When lineage matters, record command -v playwright-cli and playwright-cli --version. Only the authorised npx fallback is package-pinned; --help proves available commands, not version identity.

Skill path (set once)

export AGENTS_HOME="${AGENTS_HOME:-$HOME/.agents}"
export PWCLI="$AGENTS_HOME/skills/playwright/scripts/playwright_cli.sh"

Core workflow

Open the page.
Snapshot to get stable element refs.
Interact using refs from the latest snapshot.
Re-snapshot after navigation or significant DOM changes.
Capture only assigned artifacts; close the run-owned session.

Minimal loop:

"$PWCLI" open https://example.com
"$PWCLI" snapshot
"$PWCLI" click e3
"$PWCLI" snapshot

References

Open only what you need:

CLI command reference: references/cli.md
Practical workflows and troubleshooting: references/workflows.md

Guardrails

Before submit, send, purchase, delete, account/config mutation, upload or any other external effect, resolve action-specific authority and target. Opening or inspecting a page does not authorise the effect.
Never place real secrets in command arguments. Inventory whether the session is signed in; prefer synthetic accounts/data and an isolated named session.
Screenshots, snapshots, downloads and traces may contain credentials, personal data, DOM state, console/network content or tokens. Write only to an assigned run path, declare retention, and remove only proven run-owned artifacts.
Always snapshot before referencing element ids like e12.
Re-snapshot when refs seem stale.
Prefer explicit commands over eval and run-code unless needed.
When you do not have a fresh snapshot, use placeholder refs like eX and say why; do not bypass refs with run-code.
Use --headed when a visual check will help.
Close run-owned sessions and report any session/artifact retained for handoff.
Default to CLI commands and workflows, not Playwright test specs.