Edge Split Screen is a two-pane document multiplexer embedded inside a single browser window.
This makes Split Screen a lightweight, zero-overhead alternative to external tiling window managers.
Routing determines which pane receives new content when a link is opened.
Right-click any link to choose:
Via the Split Screen menu:
This creates a predictable flow direction for research, debugging, or documentation tasks.
Swap panes — Split Screen menu → Swap panes. Inverts left/right without altering tab order.
Rebind pane to a different tab — click inside a pane to set focus, then select any tab. That tab becomes the pane’s content.
Drag-and-drop — drag a tab into a pane’s top region. Works best with slow, deliberate placement.
Ctrl + Tab — next tabCtrl + Shift + Tab — previous tabF6 — cycles: address bar → left pane → right pane → toolbarShift + F6 — reverse cycleCtrl + Enter — open URL in same paneAlt + Enter — open in new tab, then bind to paneThese shortcuts form the speed layer for high-velocity research workflows.
Left pane holds search results, a table of contents, or an index. Right pane loads target pages. Set routing to Open links from left to right.
Best for: API docs, RFC navigation, literature review. The left pane acts as a stable anchor; the right pane is the volatile exploration surface.
Left pane: Source A. Right pane: Source B. Use Swap panes and F6 focus cycling for zero-latency cross-checking.
Best for: spec comparison, vendor evaluation, policy diffing.
Left pane: authoring surface (Notion, Obsidian, Docs). Right pane: research material. Set routing to Open links in same pane to keep the writing surface stable.
Best for: architecture docs, proposals, requirements drafting.
Left pane: logs, console, dashboards (pinned). Right pane: docs, code, issue threads (cycled with Ctrl + Tab).
Best for: cloud debugging, CI/CD failures, API troubleshooting. Persistent telemetry on the left; rotating reference context on the right.
F6 mastery eliminates the majority of mouse travelEdge Split Screen is optimized for two-stream cognition. It is the wrong tool when: