fix(tap): Detect switch-like controls for coordinate taps

CHANGELOG.md

@@ -5,6 +5,12 @@ All notable changes to the AXe iOS testing framework will be documented in this
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [Unreleased]
+
+### Fixed
+
+- Fixed coordinate taps (`tap -x/-y`) silently no-oping on SwiftUI `Toggle` controls while reporting success; with the default `automatic` tap style, AXe now inspects the element at the tapped point and uses physical touch when it reports a switch-like role. `UISwitch` internals report as plain groups at the tapped point, so coordinate taps on them still need `--tap-style physical` (or a selector tap).
+
 ## [v1.8.0] - 2026-07-20
 
 ### Added

Skills/CLI/axe/SKILL.md

@@ -7,7 +7,7 @@ description: Provides agent-ready AXe CLI usage guidance for iOS Simulator autom
 1. Identify simulator UDID target first (`axe list-simulators`).
 2. Simulator-interaction AXe commands require `--udid <UDID>`. Commands like `list-simulators` and `init` do not.
 3. Run `axe describe-ui --udid <UDID>` to inspect the full current screen. Use `axe describe-ui --point <X,Y> --udid <UDID>` to inspect the element at a specific coordinate. Use the output to discover available `--id` and `--label` values for selector taps and slider setting, and to confirm coordinates for coordinate-based taps.
-4. Prefer selectors (`tap --id` / `tap --label`, `slider --id` / `slider --label`) over raw coordinates. Selectors are resilient to layout changes, work across device sizes, and support element waiting where documented. For UIKit `UISwitch` and SwiftUI `Toggle` rows, selector taps activate the contained switch/toggle when the match contains exactly one such control. Default tap style is `automatic`: switches/toggles use physical touch down/up, while normal taps use simulator `tapAt`.
+4. Prefer selectors (`tap --id` / `tap --label`, `slider --id` / `slider --label`) over raw coordinates. Selectors are resilient to layout changes, work across device sizes, and support element waiting where documented. For UIKit `UISwitch` and SwiftUI `Toggle` rows, selector taps activate the contained switch/toggle when the match contains exactly one such control. Default tap style is `automatic`: switches/toggles use physical touch down/up, while normal taps use simulator `tapAt`. Coordinate taps (`-x`/`-y`) also inspect the element at the point when the style is `automatic`, so taps on elements that report a switch/toggle role (such as SwiftUI `Toggle`) use physical touch; pass `--tap-style simulator` to skip the inspection or `--tap-style physical` to force physical touch. UIKit `UISwitch` internals report as plain groups at the tapped point, so coordinate taps on them still need `--tap-style physical` or a selector tap.
 
 ## Step 2: Choose the right command
 

Sources/AXe/Commands/Tap.swift

@@ -106,7 +106,13 @@ struct Tap: AsyncParsableCommand {
         let resolvedDescription: String
 
         if let pointX, let pointY {
-            resolution = TapResolution(point: (x: pointX, y: pointY), isSwitchLikeControl: false)
+            resolution = await Self.resolveCoordinateTap(
+                x: pointX,
+                y: pointY,
+                requestedStyle: tapStyle ?? .automatic,
+                simulatorUDID: simulatorUDID,
+                logger: logger
+            )
             resolvedDescription = "(\(pointX), \(pointY))"
         } else {
             let query: AccessibilityQuery
@@ -176,6 +182,32 @@ struct Tap: AsyncParsableCommand {
         print("✓ Tap at \(resolvedDescription) completed successfully")
     }
 
+    @MainActor
+    static func resolveCoordinateTap(
+        x: Double,
+        y: Double,
+        requestedStyle: TapStyle,
+        simulatorUDID: String,
+        logger: AxeLogger
+    ) async -> TapResolution {
+        guard requestedStyle == .automatic else {
+            return TapResolution(point: (x: x, y: y), isSwitchLikeControl: false)
+        }
+        let isSwitchLikeControl: Bool
+        do {
+            let element = try await AccessibilityFetcher.fetchAccessibilityElement(
+                at: AccessibilityPoint(x: x, y: y),
+                simulatorUDID: simulatorUDID,
+                logger: logger
+            )
+            isSwitchLikeControl = element?.isSwitchLikeControl ?? false
+        } catch {
+            logger.info().log("Could not inspect the element at (\(x), \(y)) for automatic tap style; using simulator tap: \(error.localizedDescription)")
+            isSwitchLikeControl = false
+        }
+        return TapResolution(point: (x: x, y: y), isSwitchLikeControl: isSwitchLikeControl)
+    }
+
     private func resolvedTapStyle(for resolution: TapResolution) -> TapStyle {
         switch tapStyle ?? .automatic {
         case .automatic:

Sources/AXe/Resources/skills/axe/SKILL.md

@@ -7,7 +7,7 @@ description: Provides agent-ready AXe CLI usage guidance for iOS Simulator autom
 1. Identify simulator UDID target first (`axe list-simulators`).
 2. Simulator-interaction AXe commands require `--udid <UDID>`. Commands like `list-simulators` and `init` do not.
 3. Run `axe describe-ui --udid <UDID>` to inspect the full current screen. Use `axe describe-ui --point <X,Y> --udid <UDID>` to inspect the element at a specific coordinate. Use the output to discover available `--id` and `--label` values for selector taps and slider setting, and to confirm coordinates for coordinate-based taps.
-4. Prefer selectors (`tap --id` / `tap --label`, `slider --id` / `slider --label`) over raw coordinates. Selectors are resilient to layout changes, work across device sizes, and support element waiting where documented. For UIKit `UISwitch` and SwiftUI `Toggle` rows, selector taps activate the contained switch/toggle when the match contains exactly one such control. Default tap style is `automatic`: switches/toggles use physical touch down/up, while normal taps use simulator `tapAt`.
+4. Prefer selectors (`tap --id` / `tap --label`, `slider --id` / `slider --label`) over raw coordinates. Selectors are resilient to layout changes, work across device sizes, and support element waiting where documented. For UIKit `UISwitch` and SwiftUI `Toggle` rows, selector taps activate the contained switch/toggle when the match contains exactly one such control. Default tap style is `automatic`: switches/toggles use physical touch down/up, while normal taps use simulator `tapAt`. Coordinate taps (`-x`/`-y`) also inspect the element at the point when the style is `automatic`, so taps on elements that report a switch/toggle role (such as SwiftUI `Toggle`) use physical touch; pass `--tap-style simulator` to skip the inspection or `--tap-style physical` to force physical touch. UIKit `UISwitch` internals report as plain groups at the tapped point, so coordinate taps on them still need `--tap-style physical` or a selector tap.
 
 ## Step 2: Choose the right command
 

Sources/AXe/Utilities/AccessibilityFetcher.swift

@@ -55,6 +55,31 @@ struct AccessibilityFetcher {
         }
     }
 
+    static func fetchAccessibilityElement(
+        at point: AccessibilityPoint,
+        simulatorUDID: String,
+        logger: AxeLogger
+    ) async throws -> AccessibilityElement? {
+        let jsonData = try await fetchAccessibilityInfoJSONData(
+            for: simulatorUDID,
+            point: point,
+            logger: logger
+        )
+        return decodeSingleAccessibilityElement(from: jsonData)
+    }
+
+    static func decodeSingleAccessibilityElement(from data: Data) -> AccessibilityElement? {
+        let decoder = JSONDecoder()
+        if let element = try? decoder.decode(AccessibilityElement.self, from: data) {
+            return element
+        }
+        if let elements = try? decoder.decode([AccessibilityElement].self, from: data),
+           elements.count == 1 {
+            return elements.first
+        }
+        return nil
+    }
+
     private static func fetchAccessibilityInfoJSONData(
         from target: FBSimulator,
         at point: AccessibilityPoint

Sources/AXe/Utilities/Batch/Command+BatchConvertible.swift

@@ -80,7 +80,13 @@ extension Tap: BatchConvertible {
         let resolvedRoots: [AccessibilityElement]?
 
         if let pointX, let pointY {
-            resolution = TapResolution(point: (x: pointX, y: pointY), isSwitchLikeControl: false)
+            resolution = await Tap.resolveCoordinateTap(
+                x: pointX,
+                y: pointY,
+                requestedStyle: tapStyle ?? context.tapStyle,
+                simulatorUDID: context.simulatorUDID,
+                logger: logger
+            )
             resolvedRoots = nil
         } else {
             let query: AccessibilityQuery

Tests/AccessibilityFetcherTests.swift

@@ -271,6 +271,27 @@ struct AccessibilityFetcherTests {
         }
     }
 
+    @Test("Decodes single point-lookup elements from dictionary and singleton array payloads")
+    func decodesSinglePointLookupElements() throws {
+        let toggle: [String: Any] = [
+            "type": "CheckBox",
+            "role": "AXCheckBox",
+            "role_description": "switch",
+            "AXValue": "0",
+            "frame": ["x": 309, "y": 220, "width": 63, "height": 28],
+        ]
+        let dictionaryData = try JSONSerialization.data(withJSONObject: toggle)
+        let singletonArrayData = try JSONSerialization.data(withJSONObject: [toggle])
+        let multipleData = try JSONSerialization.data(withJSONObject: [toggle, toggle])
+
+        let fromDictionary = try #require(AccessibilityFetcher.decodeSingleAccessibilityElement(from: dictionaryData))
+        let fromArray = try #require(AccessibilityFetcher.decodeSingleAccessibilityElement(from: singletonArrayData))
+
+        #expect(fromDictionary.isSwitchLikeControl)
+        #expect(fromArray.isSwitchLikeControl)
+        #expect(AccessibilityFetcher.decodeSingleAccessibilityElement(from: multipleData) == nil)
+    }
+
     @Test("Restarts the canonical testmanagerd service with direct simctl arguments")
     func restartsCanonicalTestManagerService() async throws {
         var executableURL: URL?