Browser Automation task routing
Read this plugin only when its listed skills match the current task. Discover the actual client tools, account access and permission requirements before invoking anything. Treat this directory as bundled source, not proof of an installed or authenticated integration.
Select one entrypoint
| Skill | Apply it to |
|---|---|
| playwright | Use when the task requires automating a real browser from the terminal (navigation, form filling, snapshots, screenshots, data extraction, UI-flow debugging) via playwright-cli or the bundled wrapper script. |
| playwright-interactive | Inspect and test browser interfaces through available Playwright or Chrome DevTools MCP tools. |
| screenshot | Use when the user explicitly asks for a desktop or system screenshot (full screen, specific app or window, or a pixel region), or when tool-specific capture capabilities are unavailable and an OS-level capture is needed. |
Preserve the integration boundary
Read only the references required by the selected skill. Inspect bundled scripts before execution; retain their argument and output contracts. Keep credentials and private payloads out of examples, logs and ambient hook context.
Do not install, enable, trust, publish or update a remote integration merely because you edit these files. Keep canonical and authorized bundled mirrors coherent, preserve existing resource paths and license notices, and report the checks you actually performed. Local packaging does not assert vendor endorsement.