# AGENTS.md instructions for Playwright CLI for browser automation and test generation

Use Playwright CLI for browser automation and test generation only for: Run, debug, record, and inspect browser tests from terminal workflows.

## Default behavior

- Prefer non-interactive commands and stable output formats: json, html, junit.
- Start with read-only inspection commands before changing state.
- Show the exact command before destructive or deployment actions.
- Require explicit user approval for commands marked `requires-review` or `destructive`.

## Useful commands

- safe: `playwright@latest --help` — Inspect command surface before automation.
- safe: `playwright@latest --version` — Verify the installed CLI.

## Safety notes

- Prefer non-interactive usage and explicit output flags when an agent runs this command.
- Treat commands that deploy, delete, charge money, or mutate production data as approval-required.
- Record the exact command, output format, and scope before adding it to an agent workflow.