$ loading_
Automate a real browser from the terminal for testing, scraping, and UI debugging.
The material describes an open-source, widely adopted browser automation skill with no required secrets and no declared fixed remote endpoints, indicating overall low risk. Its inherent capabilities still include driving a real browser to access websites, potentially fetching CLI dependencies, and reading page content or creating screenshots, so it is best treated as low to moderate-low risk with normal operational caution.
The material explicitly states that no keys or environment variables are required; there is no instruction to provide API tokens, account credentials, or store secrets in config, so credential exposure appears limited.
This skill drives a real browser to visit user-specified websites and uses npx/package mechanisms to run @playwright/cli, which normally creates network connections to target sites and likely package registries. No extra third-party data collection endpoint is declared, but page contents, form inputs, and screenshots may be sent to external sites as part of normal browsing.
The README explicitly includes terminal commands, a wrapper script, and npx/npm invocations, indicating it starts local CLI/browser processes and performs automation. This is inherent to browser automation tools, and there is no evidence of requesting elevated system privileges beyond the stated purpose.
By design it can open pages, capture snapshots, extract data, and generate screenshots/traces/PDFs, which means it can read page content within the browser session and save local artifacts. The material does not show requests for broad filesystem access or unrelated data permissions, but users should assume it can access data exposed during the automated flow.
The source is an open-source GitHub repository under openai/skills with strong community adoption (about 22k stars), providing good auditability and source credibility that materially lowers supply-chain risk. Although the license and maintenance status are not explicit and the npx/npm dependency chain still merits normal scrutiny, there are no red flags such as closed source, obscure distribution channels, or obvious injection signs in the material.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "playwright" skill from askskill: 1. Download https://raw.githubusercontent.com/openai/skills/main/skills/.curated/playwright/SKILL.md 2. Save it as ~/.claude/skills/playwright/SKILL.md 3. Reload skills and tell me it's ready
Drive a real browser from the terminal using playwright-cli. Prefer the bundled wrapper script so the CLI works even when it is not globally installed.
Treat this skill as CLI-first automation. Do not pivot to @playwright/test unless the user explicitly asks for test files.
Before proposing commands, check whether npx is available (the wrapper depends on it):
command -v npx >/dev/null 2>&1
If it is not available, pause and ask the user to install Node.js/npm (which provides npx). Provide these steps verbatim:
# Verify Node/npm are installed
node --version
npm --version
# If missing, install Node.js/npm, then:
npm install -g @playwright/cli@latest
playwright-cli --help
Once npx is present, proceed with the wrapper script. A global install of playwright-cli is optional.
export CODEX_HOME="${CODEX_HOME:-$HOME/.codex}"
export PWCLI="$CODEX_HOME/skills/playwright/scripts/playwright_cli.sh"
User-scoped skills install under $CODEX_HOME/skills (default: ~/.codex/skills).
Use the wrapper script:
"$PWCLI" open https://playwright.dev --headed
"$PWCLI" snapshot
"$PWCLI" click e15
"$PWCLI" type "Playwright"
"$PWCLI" press Enter
"$PWCLI" screenshot
If the user prefers a global install, this is also valid:
npm install -g @playwright/cli@latest
playwright-cli --help
Minimal loop:
"$PWCLI" open https://example.com
"$PWCLI" snapshot
"$PWCLI" click e3
"$PWCLI" snapshot
Snapshot again after:
Refs can go stale. When a command fails due to a missing ref, snapshot again.
"$PWCLI" open https://example.com/form
"$PWCLI" snapshot
"$PWCLI" fill e1 "[email protected]"
"$PWCLI" fill e2 "password123"
"$PWCLI" click e3
"$PWCLI" snapshot
"$PWCLI" open https://example.com --headed
"$PWCLI" tracing-start
# ...interactions...
"$PWCLI" tracing-stop
"$PWCLI" tab-new https://example.com
"$PWCLI" tab-list
"$PWCLI" tab-select 0
"$PWCLI" snapshot
The wrapper script uses npx --package @playwright/cli playwright-cli so the CLI can run without a global install:
"$PWCLI" --help
Prefer the wrapper unless the repository already standardizes on a global install.
Open only what you need:
references/cli.mdreferences/workflows.mde12.eval and run-code unless needed.eX and say why; do not bypass refs with run-code.--headed when a visual check will help.output/playwright/ and avoid introducing new top-level artifact folders.Generate or edit bitmap images for illustrations, photos, textures, mockups, and cutouts.
Build, scaffold, refactor, and troubleshoot ChatGPT Apps SDK applications.
Create a composable CLI from docs, specs, SDKs, apps, or scripts.
Turn vague intentions into concrete, measurable goals and clear success criteria.
Create a new blank Figma design or FigJam file to start collaborating quickly.
Connect to Figma, fetch design assets, and turn nodes into production code.
Automate browser interactions, capture screenshots, and run end-to-end testing workflows.
Automate browser tasks with Playwright for testing, screenshots, forms, and scripting.
Automate web page workflows, login checks, tab handling, and recovery steps.
Let AI control a browser for web tasks, testing, and data extraction.
Automate browser tasks with Playwright for testing, extraction, interaction, and control.
Automate browser actions and smoke tests with screenshots, network capture, and page interaction.