Drive a native macOS app via the cua-driver MCP server or CLI โ snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. "open a file in TextEdit", "navigate to /Applications in Finder", "click the Save button in Numbers").
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
Drive a native macOS app via the cua-driver MCP server or CLI โ snapshot its AX tree, click/type/scroll by element_index, verify via re-snapshot. Use when the user asks you to operate, drive, automate, or perform a GUI task in a real macOS application on the host (e.g. "open a file in TextEdit", "navigate to /Applications in Finder", "click the Save button in Numbers").
platforms
["darwin"]
metadata
{"deepchatFeature":"computer-use"}
cua-driver
Orchestrates macOS app automation via cua-driver. Whenever a user
asks to drive a native macOS app, follow the loop in this skill rather
than calling tools ad-hoc โ the snapshot-before-action invariant is not
optional and silently breaks if you skip it.
DeepChat MCP mode
When DeepChat auto-pins this skill because the Computer Use MCP server is
enabled, use the available Computer Use tools in the current tool list
directly. Treat examples such as click({...}) or
get_window_state({...}) as calls to the matching MCP tool with the same
JSON payload. Shell CLI examples are diagnostic references for
shell-based agents.
App-name resolution belongs here. When matching a user request to an
app, compare the request with each list_apps name and bundle_id.
Consider the user's language, system language, English product names,
English brand names, romanized or pinyin variants, and common
abbreviations. Treat bundle_id as the strongest identity signal. When
the requested app name is ambiguous, localized, translated, abbreviated,
or written in another script, call , resolve the most credible
bundle id, then call with that bundle id.
list_apps
launch_app
Sparse visible UI fallback
Many media and Electron apps expose a shallow AX tree while still
showing enough pixels to act. Spotify-style snapshots with only
AXWindow, title-bar buttons, and a menu bar are still actionable when
the screenshot shows a visible, unambiguous target.
For a user-requested non-destructive action, continue within the same
turn through the safest visible options:
Re-snapshot once if the first tree looks sparse.
Use zoom on the relevant screenshot region when the screenshot is
wide, dense, or the target is small.
Prefer in-window controls visible in the screenshot: cards, primary
buttons, bottom player controls, search fields, sidebars.
Use pixel click({pid, window_id, x, y}) or click({pid, window_id, x, y, from_zoom: true}) for visible targets missing from AX.
Re-snapshot after each action and look for evidence: selected state,
text changes, playback progress, button icon changes, new panels,
highlighted rows, or changed window content.
If the first safe click has no observable effect, try the next
visible, semantically related target before asking the user.
Ask for user confirmation only when the visible candidates are ambiguous,
the action is destructive, the target is off-screen, or the next step
requires foregrounding the app. When the AX tree is weak and the
screenshot presents clear controls, continue with the visual fallback
steps above.
Menu-bar fallback is valid when the target app is already frontmost and
the menu item names match the user's requested action. Inspect menus
such as File, Edit, View, Playback, Window, and Help for semantic
commands, then dispatch menu actions and re-snapshot. For backgrounded
apps, prefer in-window AX or pixel actions.
The no-foreground contract โ read this first
The user's frontmost app MUST NOT change. This is the whole
reason cua-driver exists. Users pay for the right to keep typing in
their editor while an agent drives another app in the background.
Violate this rule and every other nice property the driver gives
you (no cursor warp, no Space switch, no window raise) stops
mattering โ you just shipped the Accessibility Inspector with extra
steps.
Before running any shell command, ask: "does this raise,
activate, foreground, or make-key any app?" If yes, don't run it.
Every one of the commands below activates the target on macOS and
is therefore forbidden unless the user explicitly asked for
frontmost state:
Every form of the open CLI โ open -a <App>, open -b <bundle-id>, open <file>, open <path-to-App.app>, open <url> โ always activates. macOS routes all forms through
LaunchServices, which unhides and foregrounds the target
regardless of whether you passed an app name, a bundle id, a
document, a URL, or the bundle path itself. The activation
happens even when the only intent was "start the process."
Never use open for any app launch. This includes launching
a just-built .app from a local build dir (e.g. open build/Build/Products/Debug/MyApp.app) โ resolve the
CFBundleIdentifier from Info.plist and use launch_app
with that id. See "The narrow carve-out" below for why
launch_app is safe even when the app internally calls
NSApp.activate.
osascript -e 'tell application "X" to activate' โ
activates by design. Same for ... to open <file>,
... to launch, and anything with activate in the tell block.
osascript -e 'tell application "System Events" to ... frontmost'
in a mutating form (setting frontmost rather than reading it).
AppleScript files that invoke activate, launch, or open
against the target app.
cliclick (moves the user's real cursor to the target coords
before clicking โ a focus-steal-equivalent even if the app's
window state is unchanged).
CGEventPost with cghidEventTap targeting a coordinate over
a different app's window (warps the cursor, possibly activates
on hit).
AppleScriptTask, NSAppleScript, Process wrapping osascript
that contains any of the above.
NSRunningApplication.activate(options:) called from your own
helper binary โ same class.
Dock clicks and any open invocation (see the first bullet โ
every form of open goes through LaunchServices which
activates, full stop).
Keyboard shortcuts that semantically mean "focus here" โ
most notably Chrome / Safari / Arc's โL (focus omnibox) and
Finder's โโงG (Go to Folder). These aren't pure key events โ
the receiving app interprets "user wants to type here" as
activation intent and raises its window to be key. Even when
delivered to a backgrounded pid via hotkey, the downstream app
pulls focus. For omnibox navigation specifically, the correct
path is launch_app({bundle_id: "com.google.Chrome", urls: ["https://โฆ"]}) โ no omnibox dance, no โL, no focus-steal. Do
NOT try set_value on the omnibox: Chrome's commit logic requires
a "user-typed" signal that neither an AX value write nor
CGEvent.postToPid keystrokes supply from a backgrounded pid โ
the URL lands in the field but Return fires as a no-op. See
WEB_APPS.md โ "Navigate to a URL" for the full pattern. The
general principle: a shortcut that says "put my cursor inside this
app" is a focus-steal; a shortcut that says "do this thing" (copy,
save, quit) is fine.
Tab-switching shortcuts in browsers (โ1..โ9, โ], โ[,
โโง[, โโง]) are visibly disruptive even when delivered to a
backgrounded pid. The app's key handler processes the shortcut,
the window re-renders the new tab's content, the user sees their
tabs flipping. There is no AX-only workaround: page content (HTML,
form state, AXWebArea) populates only for the focused tab;
inspecting a background tab requires activating it, which is the
visible flip. Observed with Dia; the same mechanic applies to every
Chromium-family browser (Chrome, Arc, Brave, Edge).
Prefer the windows-over-tabs pattern: for each URL you need to
drive backgrounded, use launch_app({bundle_id, urls: [url]}) โ
browsers open each URL in a new window. Each window has its own
window_id, its own AX tree, and can be inspected / interacted with
via element_index without activating or switching anything. Tabs
are a UX grouping for humans; cua-driver workflows should default to
windows. See WEB_APPS.md โ "Tabs vs windows" for the full pattern.
Tab-title enumeration (read-only) IS safe โ walk a window's toolbar
AX tree for AXTab / AXRadioButton children and read their
AXTitles. Tab switching (activating one) is not.
Reading frontmost state is fine (osascript -e 'tell application "System Events" to get name of first application process whose frontmost is true'). Mutating it is not.
Corollary โ the AXMenuBar rule.AXMenuBarItem + AXPick
dispatches at the AX layer regardless of which app is frontmost,
but macOS's on-screen menu bar always belongs to the frontmost
app. If you drive a backgrounded app's menu bar, the AX call
succeeds but the viewer sees the dispatch rendered over the
frontmost app's menu bar โ confusing in any observed session and
routinely a silent no-op too, because action menu items go
DISABLED when their owning app isn't the key window. So: only
use menu-bar navigation when the target is already frontmost. For
backgrounded targets, read state via in-window AX (window title,
toolbar AXStaticText) and dispatch via in-window element_index
or pixel clicks โ both paths are frontmost-insensitive. Full
rationale in "Navigating native menu bars" below.
"Open <app>" in user speech means launch, not activate.cua-driver launch_app is the one correct path for process
startup โ it's idempotent (no-op on a running app), returns the
pid, and has an internal FocusRestoreGuard that catches
NSApp.activate(ignoringOtherApps:) calls the target makes during
application(_:open:) and clobbers the frontmost back to what it
was before the launch. That guard is why launch_app with urls
(e.g. {"bundle_id": "com.colliderli.iina", "urls": ["~/video.mp4"]})
is safe even for apps that normally foreground on media-load
(Chrome, Electron, media players).
Defaults โ always prefer cua-driver over shell shims
Default transport is the cua-driver CLI โ Bash shelling out
to cua-driver <tool-name> '<JSON-args>'. MCP tools (prefix
mcp__cua-driver__*) only when the user explicitly asks for them.
CLI wins because it picks up rebuilds instantly, failures are
easier to diagnose, and there's no per-tool schema-load overhead.
Every reference to click(...), get_window_state(...) etc. in this
skill means cua-driver click '{...}' โ translate to MCP form only
when MCP is requested.
Intent โ tool mapping. If you find yourself reaching for the right
column, something has gone wrong โ re-read "The no-foreground
contract" above:
Intent
Use
Don't use
Open / launch an app
launch_app({bundle_id}) or launch_app({bundle_id, urls:[...]})
open -a, osascript 'tell app โฆ to launch/activate/open'
Find a pid
list_apps or launch_app's return
pgrep, ps, osascript frontmost
Enumerate an app's windows
list_windows({pid}) โ or read the windows array launch_app already returns
osascript 'every window of app โฆ'
Click / type / scroll / keys
click, type_text, scroll, press_key, hotkey
osascript, cliclick, raw CGEvent, open <url>
Drag / drag-and-drop / marquee select
drag({pid, from_x, from_y, to_x, to_y}) (pixel-only โ macOS AX has no semantic drag)
cliclick dd:, osascript drag
Screenshot
screenshot or the PNG in get_window_state
screencapture
Quit an app
ask the user first, then hotkey({pid, keys:["cmd","q"]})
kill, killall, pkill
Hand a file/URL to an app
launch_app({bundle_id, urls:[<path>]})
open -a <App> <path>, open <url>
The narrow carve-out
The only legitimate use of osascript -e 'tell app X to activate' is when the user explicitly asked for frontmost
state ("bring Chrome to the front", "make it frontmost", "I want
to see X"). Reaching for it because a tool call returned something
confusing is wrong โ that's the skill's classic foot-in-the-door
failure mode and it steals focus every time.
When a cua-driver call surprises you, diagnose cua-driver first:
Tiny screenshot / empty tree_markdown? Check
cua-driver get_config โ capture_mode. Default "vision" omits
the AX tree (PNG only), "ax" omits the PNG, "som" returns
both. If a snapshot lacks a tree, capture_mode is almost
certainly "vision" โ either reason purely from the PNG or flip
to "som" / "ax" via set_config.
has_screenshot: false? The window capture failed (transient
race against a close, or the window has no backing store yet).
Re-snapshot; if persistent, pick a different window_id via
list_windows.
Invalid element_index / No cached AX state? You either
skipped get_window_state this turn or passed a different
window_id than the one the snapshot cached against. The cache
is keyed on (pid, window_id) โ indices don't carry across
windows of the same app. Re-snapshot with the same window_id
you're about to click in.
Sparse Chromium AX tree? Retry get_window_state once โ the
tree populates on second call.
Only after those are ruled out, and only if the user's action
genuinely needs frontmost state, fall through to the activate
fallback. Always name the focus steal in your response ("I'll
briefly bring Chrome to the front because โฆ").
Self-check pattern
Before every Bash call whose command line touches any macOS app
(launching, opening, clicking, typing, scripting, screenshotting),
run the self-check:
Does this command foreground the target? If yes โ stop and
translate to the cua-driver equivalent from the mapping table.
Does this command move the user's real cursor? (cliclick,
any CGEventPost at cghidEventTap over another app's window).
If yes โ stop; use click({pid, x, y}) which routes per-pid
via SkyLight and never warps the cursor.
Does this command bypass cua-driver entirely? (osascript
mutating GUI state, AppleScript files, external helpers.) If
yes โ stop; find the cua-driver tool that does the intent.
If all three are "no," the command is safe. If you can't answer,
default to stop and ask rather than proceed. A single open -a
run by accident kills the demo, the trust, and the user's in-flight
editor state.
Prerequisites โ check before starting
cua-driver is on $PATH (which cua-driver). If not, point the
user at scripts/install-local.sh and stop.
Run cua-driver check_permissions (with the daemon up โ see step 3).
The default behavior also raises the system permission dialogs for
any missing grants, so the user can grant on the spot. If either
grant still reads false after that (user dismissed the dialog),
tell them to open System Settings โ Privacy & Security and grant
Accessibility and Screen Recording to DeepChat Computer Use.app, then stop.
Pass '{"prompt":false}' for a purely read-only status check that
won't steal focus.
Start the daemon with open -n -g -a "DeepChat Computer Use" --args serve (the
recommended form โ goes through LaunchServices so TCC attributes
the process to DeepChat Computer Use.app). cua-driver serve & also works;
the CLI auto-relaunches through open -n -g -a "DeepChat Computer Use" when it
detects a wrong-TCC context (any IDE-spawned shell: Claude Code,
Cursor, VS Code, Conductor). Verify with cua-driver status.
Using cua-driver from the shell
Tool names are snake_case, management subcommands are
kebab-case โ no ambiguity. Tools invoked as cua-driver <tool-name> '<JSON-args>'. Management subcommands:
open -n -g -a "DeepChat Computer Use" --args serve โ start persistent daemon
(required for element_index workflows; without it each CLI
invocation spawns a fresh process and the per-pid element cache
dies between calls). cua-driver serve & also works โ the CLI
auto-relaunches via open when the shell's TCC context is wrong.
Pass --no-relaunch / CUA_DRIVER_NO_RELAUNCH=1 to opt out.
cua-driver stop / status
cua-driver list-tools, describe <tool>
cua-driver recording start|stop|status โ see RECORDING.md
Visual cursor overlay for demos and screen recordings. Default:
enabled. Toggle with cua-driver set_agent_cursor_enabled '{"enabled":true|false}'. A triangle pointer Bezier-glides to each
click target, ring-ripples on landing, idle-hides after ~1.5s.
Motion knobs: set_agent_cursor_motion takes any subset of
start_handle, end_handle, arc_size, arc_flow, spring โ
tuneable at runtime, persisted to config.
Requires an AppKit runloop, which cua-driver serve / mcp
bootstraps. One-shot CLI invocations skip the overlay entirely.
The core invariant โ snapshot before AND after every action
Every action MUST be bracketed by get_window_state(pid, window_id):
Before โ the pre-action snapshot resolves the element_index
you're about to use. Indices from previous turns are stale; the
server replaces the element index map on every snapshot, keyed
on (pid, window_id). Indices from turn N don't resolve in turn
N+1, and indices from window A don't resolve against window B of
the same app. Skip this and element-indexed actions fail with
No cached AX state.
After โ the post-action snapshot verifies the action actually
landed. Without it you can't tell a silent no-op from a real
effect. The AX tree change (new value, new window, disappeared
menu, disabled button, etc.) is your evidence that the action
fired. If nothing changed, the action probably failed silently โ
say so, don't assume success.
This applies to pixel clicks too โ re-snapshot after to confirm the
click landed on the intended target.
Why window selection is the caller's job now
get_app_state used to pick a window for you via a max-area heuristic
that returned the wrong surface on apps with large off-screen utility
panels. Concrete reproducer: IINA's OpenSubtitles helper (600ร432
off-screen) out-area'd the visible 320ร240 player window, so
get_app_state(pid) screenshot'd the invisible panel and clicks landed
there silently. The new get_window_state(pid, window_id) makes the
caller name the window explicitly โ the driver validates that the
window belongs to the pid and is on the current Space, then snapshots
exactly what was asked for. Enumerate candidates via list_windows or
read the windows array launch_app already returns.
vision was renamed from screenshot โ the old name still decodes
as a deprecated alias, so an on-disk "capture_mode": "screenshot"
keeps working. Default is som so element_index clicks work the
first time a user calls get_window_state; the other modes are
opt-in when the caller specifically doesn't want one half of the
work. Note the tool named screenshot is separate (raw PNG, no AX
walk) and unrelated to the capture mode.
When a snapshot looks wrong (tiny screenshot / empty tree), check
cua-driver get_config for capture_mode before anything else.
Pure-vision mode has its own caveats โ Claude Code's vision
pipeline downsamples dense text aggressively, so pixel grounding
takes multiple correction cycles on text-heavy UIs. Read
SCREENSHOT.md before driving anything in that
mode; it documents the iterate/annotate/verify recipe plus the
JPEG-over-PNG finding.
Window state โ what works
state
get_window_state
click/set_value (AX)
press_key commit (Return/Space/Tab)
pixel click
frontmost
โ
โ
โ
โ
backgrounded / visible
โ
โ
โ
โ
minimized (Dock genie)
โ
โ (no deminiaturize โ AX actions fire on the minimized window in place)
โ silent no-op / system beep โ use set_value or click equivalent
โ no on-screen bounds
hidden (hides=true / NSApp.hide)
โ
โ
depends
โ
on another Space
โ ๏ธ AX tree often stripped to menu-bar-only on SwiftUI apps (System Settings) โ AppKit apps usually fine. Response carries off_space: true + window_space_ids so you can detect it
โ
โ
โ window not in current-Space list
Critical cell โ minimized + keyboard commit. The keystroke
reaches the app but AX focus doesn't propagate to renderer focus on
a minimized window. Workarounds in order of preference:
set_value to write the field's entire value directly, or AX-click
a commit-equivalent button (Go, Submit, checkbox). Tell the user
the window needs to un-minimize only as a last resort.
The canonical loop
launch_app(target)
โ pick window_id from the returned `windows` array
(or call list_windows(pid) separately)
โ get_window_state(pid, window_id)
โ [act] # every action also takes (pid, window_id)
โ get_window_state(pid, window_id) โ verify
launch_app now returns a windows array alongside the pid, so the
common case collapses to two calls (launch_app โ get_window_state)
without a separate list_windows hop.
1. Resolve target pid โ always via launch_app
Always start with launch_app, whether or not the target is already
running. It's idempotent (relaunching returns the existing pid with no
side effects) and gives you the pid in one call โ no list_apps hop.
launch_app({name: "Calculator"}) โ when bundle_id isn't known.
launch_app is a hidden-launch primitive by design โ that's the
entire point of cua-driver: agents drive apps in the background while
the user keeps typing in their real foreground app. The target's
window is initialized (AX tree fully populated, clickable via
element_index, the pid appears in list_apps) but not drawn on
screen. The driver never activates or unhides apps on its own; that
would violate the no-foreground contract the whole driver exists to
protect.
If the user explicitly wants the window visible (usually for a demo
or recording), they unhide it themselves โ Dock click, Cmd-Tab, or
Spotlight. Do not reach for open / osascript activate as a
shortcut to make the window visible; those paths break the backgrounded
invariant on every call, not just the call that "needed" the
foreground. Say out loud what the user needs to do ("click the
Todo app in your Dock to bring it forward") and let them do it.
Never shell out to any form of open (including open <path-to-App.app> for a just-built binary โ resolve the bundle id
from Info.plist and use launch_app with that), osascript 'tell app โฆ to launch/open', or similar. Those paths activate the target,
bypass the driver's focus-restore guard, and require a Bash
permission prompt the agent loop shouldn't be burning on app launch.
See "Prefer cua-driver tools over shell shims" above for the full
intent โ tool mapping.
list_apps is for app-level discovery (answering "what's installed /
running / frontmost?") โ not part of the core action loop. Skip it in
the loop. For window-level questions โ "does this app have a
visible window?", "which Space is this window on?", "which of this
pid's windows is the main one?" โ call list_windows instead; the
app record doesn't carry window state on purpose. In the common
single-window case you can skip list_windows entirely and read the
windows array that launch_app already returned.
2. Snapshot and act by element_index
Call get_window_state({pid, window_id}) with the window_id from
launch_app's windows array (or a fresh list_windows({pid}) if
you're interacting with a long-lived process). In the default
vision capture_mode the response carries only the screenshot
โ no AX tree โ so the canonical loop is list_windows โ get_window_state โ reason over PNG โ pixel click. When you need
element_index dispatch (AX-addressable elements, backgrounded
clicks), flip to som first: cua-driver set_config '{"key": "capture_mode", "value": "som"}', then call get_window_state
again. The rest of this section walks through som mode, which
is what you want once you've decided element-indexed addressing is
required.
In som mode the response carries:
tree_markdown โ every actionable element tagged [N]. That N
is the element_index. The tree can be very large (Finder is
~1600 elements, ~190 KB); when it exceeds token limits the MCP
harness saves it to a file and returns the path. Use Bash +
jq -r '.tree_markdown' + grep to pull the section you need.
screenshot_png_b64 + screenshot_width / _height /
_scale_factor โ the window screenshot (actually JPEG-85 despite
the _png_ field name, hard-coded in
WindowCapture.captureFrontmostWindow). Present in som mode
(spliced into the structured JSON alongside the tree). In vision
mode the image arrives as a native MCP image content block with no
structured wrapper. Omitted when the target has no on-screen
window.
has_screenshot: bool โ gate on this before piping the PNG.
Otherwise jq -r '.screenshot_png_b64' emits the literal
"null", base64-decodes into 3 bytes of garbage, and downstream
vision APIs reject it with an opaque "Could not process image"
error.
# canonical, works in every capture mode โ writes the image bytes
# wherever you point, stdout stays readable (tree in som, summary
# in vision). stderr warns (exit 0) if the response had no image.
cua-driver get_window_state '{"pid":N,"window_id":W}' --image-out /tmp/shot.png
# som-only legacy path: pull the spliced base64 out of structuredContent.
# Prefer --image-out above โ it's one flag vs a probe + pipe.
if [ "$(cua-driver get_window_state '{"pid":N,"window_id":W}' | jq -r '.has_screenshot')" = "true" ]; then
cua-driver get_window_state '{"pid":N,"window_id":W}' | jq -r '.screenshot_png_b64' | base64 -d > shot.png
fi
Reason over both the tree AND the screenshot โ they're
complementary, not redundant. In som mode every
turn's get_window_state gives you both halves and you should pull
signal from each:
The AX tree tells you what's clickable โ roles, labels,
element_index handles, advertised actions, parent-child
structure. This is the ground truth for dispatching.
The screenshot tells you which one โ the tree often has
many buttons with similar or empty labels ("Delete", "OK",
anonymous UUID-labeled buttons, five AXStaticText = " "), and
visual context disambiguates. Captions, colors, layout relationships
visible in pixels often don't show up in the AX tree at all
(especially in Chromium / Electron / web content).
Canonical AX pattern: look at the screenshot to decide "the blue
Subscribe button on the top-right of the video card", then walk the
tree to find the matching AXButton and dispatch by its
element_index. This is the preferred path when the target is
present in the AX tree.
Canonical visual fallback: if the requested control is visible in
the screenshot and the AX tree is sparse, generic, or missing that
control, click the window-local screenshot coordinates in the same
turn. Use click({pid, window_id, x, y}); after a zoom call use
click({pid, window_id, x, y, from_zoom: true}). Re-snapshot after
the click to verify. Continue without asking for an extra "go ahead"
when the user already requested the action and the target is visible
and unambiguous. Pause only when the target is visually ambiguous,
off-screen, destructive, or would require foregrounding the app.
The actions=[...] list on each element is advisory, not
authoritative. cua-driver does not pre-flight check against it โ
click({pid, element_index}) always attempts AXPress (or the
action you pass) and surfaces whatever the target returns. Many
apps accept AXPress on elements that don't advertise it โ Chrome's
omnibox suggestion AXMenuItem is a live example. Try the click
first โ pivot only on the returned AX error code.
Dispatch table (every row assumes a (pid, window_id) pair from the
last get_window_state; window_id is required alongside
element_index, ignored on pixel-only forms unless you want to
anchor the conversion against a specific window):
Intent
Tool
Notes
List an app's windows
list_windows({pid})
returns window_id, title, bounds, z_index, is_on_screen, on_current_space. Already included in launch_app's response โ only call this for long-lived pids
Snapshot a window
get_window_state({pid, window_id})
returns tree_markdown + screenshot_*; populates the (pid, window_id) element_index cache
Left click
click({pid, window_id, element_index})
default action: "press". Pixel form: click({pid, x, y}) (window_id optional โ when supplied, pinpoints the anchor window) โ modifier: ["cmd"]
Double-click / open
double_click({pid, window_id, element_index})
AXOpen when advertised (Finder items, openable rows); else stamped pixel double-click at the element's center. Pixel form: double_click({pid, x, y}) โ primer-gated recipe lands on backgrounded Chromium web content (YouTube fullscreen, Finder open-on-dbl). click({..., count: 2}) still works and routes through the same recipe; double_click is the intent-first spelling
Right click / context menu
right_click({pid, window_id, element_index}) or click({pid, window_id, element_index, action: "show_menu"})
Chromium web-content coerces pixel right-click to left โ see WEB_APPS.md
Type at cursor
type_text({pid, text, window_id, element_index})
AXSelectedText write; focuses first
Set whole field value
set_value({pid, window_id, element_index, value})
sliders, steppers, text fields; use for keyboard-commit workarounds on minimized windows
AX insertion with CGEvent-to-pid fallback; reaches Chromium/Electron inputs
All keyboard/text primitives require pid. There is no
frontmost-routed variant โ every key goes to the named target via
CGEvent.postToPid, so the driver cannot leak keystrokes into the
user's foreground app.
Why element_index is the primary path: works on hidden /
occluded / off-Space windows, no focus steal, stable across
rebuilds, labels tell you what you're clicking. Use pixel
coordinates as the active fallback for visible self-drawn controls,
canvas/video/WebGL surfaces, and sparse AX trees.
Pixel-coordinate clicks
The pixel path (click({pid, x, y})) is for surfaces the AX tree
doesn't reach โ canvases, video players, WebGL, custom-drawn controls.
Coords are window-local screenshot pixels (same space as the PNG
get_window_state returns). Top-left origin, y-down. The driver
handles screen-point conversion internally. Passing window_id
alongside x, y is optional but recommended โ it pins the
coordinate conversion to the window whose screenshot produced the
pixel, rather than the driver's heuristic choice.
Reading coordinates from the PNG
PNGs returned by get_window_state are capped at 1568 px
long-side by default (max_image_dimension config), matching
Anthropic's multimodal-vision downsampling limit. That means the
image the model reasons over and the image the click tool's
coordinate system lives in are the same resolution โ just look
at the PNG, pick a pixel, click at that pixel. No scaling math.
This is the default because the mismatch between "rendered
thumbnail" and "native PNG" was a recurring coord-estimation
footgun. If you opt out (explicit max_image_dimension=0 for
pixel-perfect verification flows), the old rule applies: don't
eyeball coords from whatever your client renders โ it may be
2-4ร smaller than the PNG on disk, and a 2% error in thumbnail
space becomes ~80 px in the real image. Use the crosshair recipe
below against the full-resolution file in that case.
get_window_state({pid, window_id}) returns an image capped
at 1568 long-side (default) plus its dimensions
(screenshot_width / screenshot_height). Write the bytes to
disk with --image-out <path> in any capture mode โ works
identically in vision (where it's the only way) and som
(where it sidesteps the jq + base64 dance on the spliced
screenshot_png_b64 field).
You are a multimodal model โ look at the PNG. Since the PNG
matches what you see, pick the target pixel directly. No
fractional math needed.
When precision matters (small targets, dense UIs), draw a
crosshair on the image (do not crop โ cropping loses the
coordinate system and requires error-prone offset math) and
show it before clicking:
from PIL import Image, ImageDraw
img = Image.open('/tmp/shot.png')
draw = ImageDraw.Draw(img)
x, y = <your_coordinate>
r = 18
draw.ellipse([x-r, y-r, x+r, y+r], outline='red', width=4)
draw.line([x-30, y, x+30, y], fill='red', width=3)
draw.line([x, y-30, x, y+30], fill='red', width=3)
img.save('/tmp/shot_annotated.png')
Only dispatch the click after the user (or your own re-read of
the annotated image) confirms the crosshair is on target.
Addressing variants
click({pid, x, y}) โ single left-click.
click({pid, x, y, count: 2}) โ double-click.
click({pid, x, y, modifier: ["cmd"]}) โ cmd-click. Accepts any
subset of cmd/shift/option/ctrl.
right_click({pid, x, y}) โ also takes modifier.
The pixel path animates the agent cursor overlay but never warps
the real cursor. If the pid has no on-screen window the call errors
with pid X has no on-screen window โ you need a visible window to
anchor the conversion.
How the pixel click is dispatched
The recipe is the backgrounded "noraise" sequence: yabai's
focus-without-raise SLPS event records followed by an off-screen
user-activation primer and the real click, all stamped via
SLEventPostToPid. The target app becomes AppKit-active for event
routing but its window does not rise to the front of the
z-stack, and macOS's "switch to Space with windows for app" follow
is suppressed. Full mechanics in
Sources/CuaDriverCore/Input/MouseInput.swift (clickViaAuthSignedPost)
and the companion FocusWithoutRaise.swift.
Known limits
Chromium <video> play/pause: pixel click is often rejected
by HTML5's click-to-play handler on some builds. Use keyboard
instead: press_key({pid, key: "k"}) (YouTube) or
press_key({pid, key: "space"}) (generic). Keyboard events
travel through a different auth envelope.
Pixel right-click on Chromium web content coerces to a
left-click โ a known Chromium renderer-IPC limitation that affects
every non-HID-tap synthesis path. For context menus on
AX-addressable elements (links, buttons, toolbar items), use
right_click({pid, element_index}) instead.
Canvases, viewports, games (Blender, Unity, GHOST, Qt, wxWidgets)
Apps whose main surface is an OpenGL / Metal / Qt / wxWidgets
viewport expose no useful AX tree โ the whole surface is one
opaque AXGroup or AXWindow from AX's perspective. Per-pid event
paths (SLEventPostToPid, CGEvent.postToPid) are filtered by the
viewport's own event-source check and silently dropped โ the event
loop wants "real HID origin".
The working pattern:
Bring the target frontmost (a brief osascript activate is
acceptable here โ this is the carve-out the skill's osascript
gate allows).
CGEvent.post(tap: .cghidEventTap) with a leading mouseMoved
event (~30 ms before the click). cua-driver click when the
target is frontmost automatically takes this path.
Accept that the real cursor visibly moves โ cghidEventTap is
the system HID stream, the cursor warps to the click point.
There is no backgrounded path that reaches these apps today.
Navigating native menu bars (AXMenuBar)
Only drive the menu bar when the target app is frontmost. This
is the single most-misused cua-driver capability. If the target is
backgrounded, don't reach for AXMenuBarItem + AXPick โ use
in-window element_index or pixel clicks instead. Two reasons, one
functional and one perceptual:
Functional: menu items that touch document/playback/editor
state go DISABLED when their owning app isn't the key window
(Preview rotate, IINA speed change, most editor commands). AXPick
AXPress will dispatch successfully from the driver's side but
no-op at the target โ you get a silent false-pass.
Perceptual (matters for demos, screen recordings, and anything
the user watches live): macOS's screen-rendered menu bar
always belongs to the frontmost app. AXPick on a backgrounded
app's AXMenuBarItem dispatches to that app's per-process menu at
the AX layer, but any visible menu render happens over the
frontmost app's menu bar โ the viewer sees an IINA submenu
flashing on top of Chrome's menus, which reads as "the agent
clicked the wrong app." The AX call was correct; the frame the
user sees is not. For recorded or observed sessions, this is an
integrity bug even though it's not a correctness bug.
Good decision rule: if the target is not already frontmost, do
not use AXMenuBarItem at all. For reading in-window state,
snapshot the window AX tree โ most apps expose the same state via
an in-window AXStaticText, title bar, or toolbar. For dispatching
actions, use in-window element_index (buttons, toolbar items) or
pixel clicks on in-window controls โ both dispatch via AppKit's
window-under-pointer hit-test and are not frontmost-gated.
When the target IS frontmost, the menu-bar flow below is fine and
the canonical path for menus.
The two-snapshot pattern (target frontmost only)
Menu contents are a two-snapshot flow. Closed AXMenu subtrees are
deliberately skipped during snapshot โ otherwise every app's File /
Edit / View hierarchy plus every Recent Items macOS has ever seen
would inflate the tree 10-100x. But once a menu is open, its
AXMenuItem children do receive element_index values so you can
click them normally.
Find the [N] AXMenuBarItem "<Menu Name>" in the tree.
click({pid, element_index: N, action: "pick"}) โ menu bar items
implement AXPick ("open my submenu"), not AXPress. Using the
default action on an AXMenuBarItem is a no-op.
Re-snapshot. The expanded menu's items now appear under the bar
item as [M] AXMenuItem "<Item Name>".
Click the target item โ most items respond to AXPress (default
action). Submenus nest under the item and are walked the same way.
Re-snapshot and verify.
If you ever need to back out without selecting, press_key({pid, key: "escape"}) closes the open menu. Leaving a menu expanded between
turns poisons subsequent snapshots for that pid.
Commands gated on the target being frontmost
Some menu items and global shortcuts (Preview's Tools โ Rotate
Right, โR; anything in the View menu that manipulates the current
document; most editor commands) are disabled unless the target
app is the key / frontmost window. You'll see it in the AX tree
as DISABLED on the menu item even though the user's intent is
obviously valid.
Before activating, confirm you're in this narrow case โ the menu
item still reads DISABLED after a fresh snapshot AND the action
the user requested genuinely requires frontmost (Preview rotate,
View menu document manipulation, editor commands). If either
check fails, don't activate.
When both checks pass, the driver has no activate tool
(deliberately โ the whole point is backgroundable control), so
this is the one legitimate osascript fallback:
osascript -e 'tell application "<App Name>" to activate'
Then re-snapshot โ the menu item loses its DISABLED tag โ and
click({action: "pick"}) the item. Alternatively, a hotkey
call delivered to the now-frontmost app works for the shortcut
form (โR, โ+, etc.).
Always name the focus steal in your response so the user isn't
surprised โ "Briefly activating Preview to enable Tools โ Rotate
Right" or similar. Don't silently steal focus. You don't need to
restore the previous frontmost afterwards unless the user asks โ
they can cmd-tab back.
Web-rendered apps (browsers, Electron, Tauri)
For Chrome / Edge / Brave / Arc / Safari, Electron apps (Slack,
VSCode, Notion, Discord), and Tauri apps โ see WEB_APPS.md.
Covers: sparse AX tree population (retry-once pattern for Chromium),
URL navigation via omnibox suggestions, the set_value workaround
for keyboard commits on minimized windows (Return silently
no-ops โ symptom is a macOS system beep; use set_value or click a
clickable equivalent), scrolling via synthetic PageUp/Down keystrokes,
in-page clicks, and typing into web inputs.
Chromium web content specifically also coerces right_click back to
left โ use element_index for AX-addressable targets and accept the
limit otherwise.
Browser JS primitives โ page tool and get_window_state(javascript=)
When the AX tree doesn't expose the data you need (common in
Chromium/Electron โ the tree is sparse for web content), use the
page tool or the javascript param on get_window_state to query
the DOM directly via Apple Events. Requires "Allow JavaScript from
Apple Events" to be enabled โ see WEB_APPS.md for the setup path.
Three actions on the page tool:
page({pid, window_id, action: "get_text"}) โ returns
document.body.innerText. Fastest way to read page content, prices,
article text, or any raw text the AX tree truncates or omits.
page({pid, window_id, action: "query_dom", css_selector: "a[href]", attributes: ["href"]}) โ runs querySelectorAll and returns each
match's tag, text, and requested attributes as a JSON array. Use for
table rows, link hrefs, data attributes, structured page data.
page({pid, window_id, action: "execute_javascript", javascript: "..."}) โ raw JS. Wrap in an IIFE with try-catch. Don't use this for
elements already indexed by get_window_state โ click and
set_value are more reliable there.
Co-located read โ get_window_state with javascript:
Runs the JS and appends the result as a ## JavaScript result section
alongside the AX snapshot โ one round-trip instead of two. Use this
when you need both the element tree (for subsequent clicks) and some
page data in the same turn.
Decision rule โ AX vs JS:
Need
Use
Click / type into an element
get_window_state โ click / set_value (AX, works backgrounded)
Read text the AX tree drops
page(get_text) or get_window_state(javascript=)
Scrape structured data (tables, hrefs)
page(query_dom)
Trigger JS events / mutations
page(execute_javascript)
Supported backends:
App type
How
Context
Chrome / Brave / Edge
Apple Events execute javascript
Full DOM โ
Safari
Apple Events do JavaScript
Full DOM โ
Electron (VS Code, Cursorโฆ)
SIGUSR1 โ V8 inspector โ CDP
Main process only: process, Buffer โ no document, no require in sandboxed apps
Electron (with --remote-debugging-port)
CDP page target
Full DOM โ
Electron sandbox note: SIGUSR1 connects to the Node.js main process.
Sandboxed Electron apps (VS Code, Cursor) strip require and Electron
APIs there. Useful for: process.env, process.versions, process.cwd(),
process.pid. For full DOM/renderer access, launch the app with
--remote-debugging-port=9222 โ cua-driver will detect and prefer the
page target automatically.
Arc returns no values; Firefox has no JS-via-AppleEvents support โ see
WEB_APPS.md for the full matrix.
3. Re-snapshot and verify โ mandatory
Always call get_window_state({pid, window_id}) after the action.
This isn't optional verification โ it's the second half of the
snapshot invariant.
Check the AX tree diff: a changed value, a new element, a new
window, or the disappearance of the thing you just clicked (menus
collapse after selection, buttons may become disabled, etc.). If
nothing changed, the action likely failed silently โ tell the
user what you attempted and what you observed, don't paper over
with "done" language. Agents that skip this step report success on
silently-dropped actions โ the single most common failure mode.
Recording trajectories
Session-scoped action recording + replay, for demos, regressions, and
training data. Only invoke when the user explicitly asks to record a
session โ the skill does not auto-enable this. CLI surface:
cua-driver recording start|stop|status; raw tool: set_recording.
See RECORDING.md for the full flow: enable/disable, turn folder
contents, replay via replay_trajectory, and the element_index
doesn't-survive-across-sessions caveat.
Common error patterns
Error text
Meaning
Fix
No cached AX state for pid X window_id W
You either skipped get_window_state this turn, or passed a different window_id to the click than the one the snapshot cached against
Call get_window_state({pid: X, window_id: W}) first โ the same window_id you intend to click in
Invalid element_index N for pid X window_id W
Index is stale or out of range
Re-run get_window_state with the same window_id, pick a fresh index from the new tree
window_id W belongs to pid P, not โฆ
Passed a window_id that's owned by a different process
Use list_windows({pid: X}) to enumerate this pid's own windows
AX action AXPress failed with code โฆ
Element doesn't support AXPress
Try show_menu, confirm, cancel, or pick
macOS system-alert beep on press_key with no visible change
Target window is minimized; Return / Space / Tab commits don't establish real renderer focus on minimized windows
AX-click a clickable equivalent (Go button, Submit button, checkbox) instead of pressing the key; see "Keyboard commits on minimized windows" under the Browser section
Accessibility permission not granted
TCC not granted
Stop; tell user to grant in System Settings
Screen Recording permission not granted
TCC not granted for capture
Affects screenshot and get_window_state (which always captures). Grant in System Settings โ the driver can't operate without it
Things to avoid
Never reuse an element_index across a re-snapshot of the same pid.
Never translate screenshot pixels into a click โ the screenshot
is for visual disambiguation, not coordinates. Use the
element_index.
Prefer AX over pixels.click({pid, x, y}) works for
canvas / WebView regions, but it lands blindly and skips the
agent-cursor overlay. Exhaust AX paths (menu bars, cmd-k palettes,
toolbar items, keyboard shortcuts) before dropping to coordinates.
Never drive destructive actions (delete files, close unsaved
documents, send messages, submit forms) without explicit user
intent for that specific destructive step.
Never launch apps autonomously; confirm with the user first
unless their original request clearly implies the launch.
Example end-to-end task
User: "Open the Downloads folder in Finder."
launch_app({bundle_id: "com.apple.finder", urls: ["~/Downloads"]})
โ {pid: 844, windows: [{window_id: 6123, title: "Downloads", ...}]}.
Idempotent launch; plus Finder opens a hidden window rooted at
~/Downloads via application(_:open:) โ zero activation, no
focus steal. The windows array lets you skip a list_windows hop.
get_window_state({pid: 844, window_id: 6123}) โ verify an
AXWindow whose title contains "Downloads" is present with a
populated AX subtree (sidebar, list view, files).
Done.
If the user instead asks to navigate within an already-open Finder
window, use the menu-bar flow from the "Navigating native menu bars"
section above (click Go โ pick a menu item โ re-snapshot โ click it).
