Layer Command References

Playwright MCP

Tools are already in your context. Use mcp__plugin_playwright_playwright__* tools directly. Key guidance:

• Targeting: ref-based. Call browser_snapshot first to get an accessibility tree with element refs ([ref="e5"]). Use refs in browser_click, browser_type, etc.
• Refs invalidate after any state-changing action (click, type, navigate). Always re-snapshot before the next interaction.
• Tabs: browser_tabs handles list, create, select, and close.
• Wait: browser_wait_for accepts text to wait for or a timeout.
• Screenshot: browser_take_screenshot captures the current viewport.

playwright-cli

Run playwright-cli help to get the installed command list. For detailed docs, look up /microsoft/playwright-cli via Context7 (optional -- local help is sufficient if Context7 is unavailable).

Key guidance:

• Targeting: ref-based. Run playwright-cli snapshot to get element refs. Use refs with click, fill, dblclick, hover, select.
• Refs invalidate after state-changing commands. Re-snapshot before next ref-based action.
• Navigate current tab: playwright-cli goto <url>
• Open new tab: playwright-cli open <url> (background) or playwright-cli tab-new <url>
• Tabs: tab-list, tab-select <index>, tab-close
• Session history: cat /.playwright/session.md for command log recovery.

cmux-browser

Run these to get the command surface and discover browser surfaces:

cmux browser --help
cmux identify --no-caller
cmux list-pane-surfaces

If no browser surface exists, create one:

cmux new-surface --type browser --pane <ref> --url <url>

All commands follow the pattern: cmux browser --surface <ref> <subcommand>.

Key guidance:

• Targeting: selector-based. Use CSS selectors for click, fill, type.
• Surface refs are dynamic -- discover via cmux identify, never hardcode.
• No file:// URLs -- content must be served over HTTP.
• Snapshot: cmux browser --surface <ref> snapshot --compact
• Navigate: cmux browser --surface <ref> navigate <url>
• Eval: cmux browser --surface <ref> eval <expression>
• Tabs: cmux browser --surface <ref> tab new|list|switch|close
• Wait: cmux browser --surface <ref> wait --load-state complete
• Unique features: highlight <selector>, addstyle <css>, addscript <js>, state save|load <path> for checkpointing.

CDP

Store the resolved CDP_JS path. All commands use node "$CDP_JS" <command>.

Run node "$CDP_JS" (no args) to see the full command list.

Key guidance:

• Targeting: selector-based. Use CSS selectors for click, type.
• Page understanding: ax-tree is the primary method (semantic roles and names). Use dom as fallback for raw HTML.
• Screenshots: save to /tmp/, then use the Read tool to view the PNG.
• Eval: supports promises: eval "await fetch('/api').then(r=>r.json())"
• Tab targeting: use list to see tabs with IDs, then --id <target-id> on any command.
• Tab workarounds: eval "window.open('<url>')" then list for new tabs. eval "window.close()" to close (only works on script-opened tabs).
• Streaming: console and network commands stream events for debugging (not available in other layers).