Observe the desktop. Returns interactive entities with leases (UIA, CDP, Terminal, Visual SoM).
Perform actions (click, type, drag, select) on entities via lease validation. Returns semantic diffs — plus an optional `roiCapture` (changed-region PNG + next-target preview) on visual-only targets.
Lightweight check of focus, active window, cursor, and Auto-Perception attention signal.
Multi-mode capture: `detail='text'` (UIA/OCR), `diffMode` (P-frame), `dotByDot` (1:1), and `background`. Returns a cheap `screenshot://by-ref/{id}` link to the saved image instead of inlining pixels every time.
Inspect and prune the on-disk screenshot cache behind the by-ref links: `screenshot_query` lists saved captures without re-reading pixels; `screenshot_gc` reclaims space by retention policy (dry-run by default).
Instant session orientation: all window thumbnails + UI summaries in one call.
Diagnostic check for native engine health and feature activation.
Send keyboard input. Supports background input (WM_CHAR) and IME-safe clipboard bypass.
Precision coordinate-based interaction with homing and force-focus protection.
Multi-strategy: `raw` (notches), `to_element`, `smart` (virtual lists), and `capture` (stitch).
Legacy UIA-based click by name/ID (fallback when entities are unavailable).
Idempotent debug-mode launch and reliable navigation.
+11 more tools listed on main page