@swmansion/argent 0.32.1-next.17 → 0.32.1-next.19

package/dist/cli-cmds.mjs

@@ -35871,7 +35871,7 @@ var _CI_VENDOR_COUNT_FOR_TEST = vendors_default.length;
 var SESSION_ID2 = randomUUID5();
 function readCliVersion() {
   if (true) {
-    return "0.32.1-next.17";
+    return "0.32.1-next.18";
   }
   return "0.0.0";
 }

package/dist/installer.mjs

@@ -16981,7 +16981,7 @@ var _CI_VENDOR_COUNT_FOR_TEST = vendors_default.length;
 var SESSION_ID = randomUUID4();
 function readCliVersion() {
   if (true) {
-    return "0.32.1-next.17";
+    return "0.32.1-next.18";
   }
   return "0.0.0";
 }

package/dist/mcp-server.mjs

@@ -29170,6 +29170,78 @@ function toMcpToolList(tools, options = {}) {
   return tools.filter((t) => !t.hideFromMcp).map((t) => toMcpTool(t, options));
 }
 
+// ../argent-mcp/src/list-devices-text.ts
+var RUNTIME_PREFIX = "com.apple.CoreSimulator.SimRuntime.";
+var STOPPED_SIMULATORS_HEADER = "iOS simulators that are shut down, one per line (platform, udid, name, runtime, state, tags); start one with boot-device:";
+function isRecord3(value) {
+  return value !== null && typeof value === "object" && !Array.isArray(value);
+}
+function shortRuntime(runtime) {
+  if (!runtime.startsWith(RUNTIME_PREFIX)) return runtime;
+  const rest = runtime.slice(RUNTIME_PREFIX.length);
+  const match = /^([A-Za-z]+)-(\d+(?:-\d+)*)$/.exec(rest);
+  return match ? `${match[1]} ${match[2].replace(/-/g, ".")}` : rest;
+}
+function cleanEntry(device) {
+  const out = {};
+  for (const [key, value] of Object.entries(device)) {
+    if (value === null || value === void 0) continue;
+    if (key === "isEmulator") continue;
+    if (key === "runtimeKind" && value === "mobile") continue;
+    out[key] = key === "runtime" && typeof value === "string" ? shortRuntime(value) : value;
+  }
+  return out;
+}
+function isStoppedSimulator(device) {
+  if (device.platform !== "ios" && device.platform !== "ios-remote") return false;
+  if ("kind" in device || "external" in device) return false;
+  return typeof device.udid === "string" && device.state === "Shutdown";
+}
+var ONE_LINE_FIELDS = /* @__PURE__ */ new Set(["platform", "udid", "name", "runtime", "state"]);
+function quoteIfNeeded(value) {
+  return value === "" || /[\s"\\]/.test(value) ? JSON.stringify(value) : value;
+}
+function stoppedSimulatorLine(device) {
+  const d = cleanEntry(device);
+  const name = typeof d.name === "string" ? d.name : "";
+  const parts = [String(d.platform), String(d.udid), JSON.stringify(name)];
+  if (typeof d.runtime === "string") parts.push(d.runtime);
+  if (typeof d.state === "string") parts.push(quoteIfNeeded(d.state));
+  for (const [key, value] of Object.entries(d)) {
+    if (ONE_LINE_FIELDS.has(key)) continue;
+    if (value === true) parts.push(key);
+    else if (typeof value === "string" || typeof value === "number")
+      parts.push(`${key}=${quoteIfNeeded(String(value))}`);
+    else parts.push(`${key}=${JSON.stringify(value)}`);
+  }
+  return parts.join(" ");
+}
+function listDevicesText(result) {
+  if (!isRecord3(result) || !Array.isArray(result.devices)) return void 0;
+  if (!result.devices.every(isRecord3)) return void 0;
+  const devices = result.devices;
+  const listed = [];
+  const stopped = [];
+  for (const device of devices) {
+    if (isStoppedSimulator(device)) stopped.push(stoppedSimulatorLine(device));
+    else listed.push(JSON.stringify(cleanEntry(device)));
+  }
+  const rest = [];
+  for (const [key, value] of Object.entries(result)) {
+    if (key === "devices") continue;
+    rest.push(`${JSON.stringify(key)}:${JSON.stringify(value)}`);
+  }
+  const devicesPart = listed.length > 0 ? `[
+${listed.join(",\n")}
+]` : "[]";
+  const object3 = `{"devices":${devicesPart}${rest.map((r) => `,
+${r}`).join("")}}`;
+  if (stopped.length === 0) return object3;
+  return `${object3}
+${STOPPED_SIMULATORS_HEADER}
+${stopped.join("\n")}`;
+}
+
 // ../argent-mcp/src/instructions.ts
 var SERVER_INSTRUCTIONS = "Argent \u2014 iOS Simulator, Android Emulator, and Chromium app control for interacting, testing, profiling and debugging mobile and Chromium applications. Interaction tools return the screen after the action: a screenshot plus the accessibility element tree with normalized tap frames. Take coordinates from that tree; call describe only when no fresh tree is available \u2014 never guess coordinates from pixels. Do not stop simulator servers when a task finishes: the next task reuses them. Call stop-all-simulator-servers when the user asks to clean up or ends the session, or when a skill tells you to recycle a device's services, always with devices: [...] naming the devices this session used. Never stop a Metro the user started unless they ask. One tool-server is shared by every agent using this argent install, so an unscoped call tears down their devices too \u2014 reserve it for a deliberate machine-wide cleanup.";
 
@@ -29480,8 +29552,11 @@ ${hintLine}${desc}` }];
         authToken: AUTH_TOKEN,
         deviceId: getDeviceIdFromArgs(params.arguments)
       };
+      const devicesText = canonicalToolId(params.name) === "list-devices" ? listDevicesText(result) : void 0;
       let content;
-      if (canonicalToolId(params.name) === "e2e-execute" && result && typeof result === "object" && ("e2eTest" in result || "flow" in result) && "steps" in result) {
+      if (devicesText !== void 0) {
+        content = [{ type: "text", text: devicesText }];
+      } else if (canonicalToolId(params.name) === "e2e-execute" && result && typeof result === "object" && ("e2eTest" in result || "flow" in result) && "steps" in result) {
         content = await flowRunToMcpContent(result, ctx);
       } else if (params.name === "screenshot-diff" && isScreenshotDiffResult(result)) {
         content = await screenshotDiffToMcpContent(result, ctx);

package/dist/tool-args.generated.d.ts

@@ -506,7 +506,7 @@ export interface ArgentToolArgs {
         includeAncestors?: boolean;
         /** Include child views for each matched view (default true) */
         includeChildren?: boolean;
-        /** View fields to include. Defaults: className, frame, hidden, alpha, identifier, label, nativeID, userInteractionEnabled, depth. Additional: pointer, tag, windowFrame, screenFrame, bounds, center, opaque, clipsToBounds, transform, contentMode, backgroundColor, tintColor, layerName */
+        /** View fields to include. Defaults: pointer, className, tag, frame, windowFrame, screenFrame, bounds, hidden, alpha, opaque, clipsToBounds, userInteractionEnabled, depth, identifier, label, layerName, nativeID. Additional: center, transform, contentMode, backgroundColor, tintColor, value, firstResponder */
         fields?: Array<string>;
     };
     "native-full-hierarchy": {
@@ -514,7 +514,7 @@ export interface ArgentToolArgs {
         udid: string;
         /** Bundle ID of the app (on a Mac app, the bundle id in the device id) */
         bundleId: string;
-        /** View fields to include. Use EXACT names: className, frame, hidden, alpha, identifier, label, nativeID, userInteractionEnabled, depth, pointer, tag, windowFrame, screenFrame, bounds, center, opaque, clipsToBounds, transform, contentMode, backgroundColor, tintColor, layerName. Defaults to all of the first group when omitted. */
+        /** View fields to include. Use EXACT names. Defaults: pointer, className, tag, frame, windowFrame, screenFrame, bounds, hidden, alpha, opaque, clipsToBounds, userInteractionEnabled, depth, identifier, label, layerName, nativeID. Additional: center, transform, contentMode, backgroundColor, tintColor, value, firstResponder */
         fields?: Array<string>;
         /** Exact UIView class names whose entire subtree should be pruned (e.g. ["UIImageView"] to drop image leaf nodes) */
         skipClasses?: Array<string>;
@@ -560,7 +560,7 @@ export interface ArgentToolArgs {
         udid: string;
         /** Bundle ID of the app (on a Mac app, the bundle id in the device id) */
         bundleId: string;
-        /** Raw X coordinate in points, on the portrait axes of the screen (the axes of native-describe-screen frames); on a Mac app, points from the top-left of the area describe reports. NOT normalized [0,1] tap space. */
+        /** Raw X coordinate in points, on the portrait axes of the screen; on a Mac app, points from the top-left of the area describe reports. NOT normalized [0,1] tap space. */
         x: number;
         /** Raw Y coordinate in points, on the same axes as x. NOT normalized [0,1] tap space. */
         y: number;
@@ -568,7 +568,7 @@ export interface ArgentToolArgs {
         includeAncestors?: boolean;
         /** Include child views for the matched view (default false) */
         includeChildren?: boolean;
-        /** View fields to include. Defaults: pointer, className, tag, frame, windowFrame, screenFrame, bounds, hidden, alpha, opaque, clipsToBounds, userInteractionEnabled, depth, identifier, label, layerName, nativeID. Additional: center, transform, contentMode, backgroundColor, tintColor */
+        /** View fields to include. Defaults: pointer, className, tag, frame, windowFrame, screenFrame, bounds, hidden, alpha, opaque, clipsToBounds, userInteractionEnabled, depth, identifier, label, layerName, nativeID. Additional: center, transform, contentMode, backgroundColor, tintColor, value, firstResponder */
         fields?: Array<string>;
         /** Exact UIView class names whose entire subtree should be pruned */
         skipClasses?: Array<string>;
@@ -582,7 +582,7 @@ export interface ArgentToolArgs {
         udid: string;
         /** Bundle ID of the app (on a Mac app, the bundle id in the device id) */
         bundleId: string;
-        /** Raw X coordinate in points, on the portrait axes of the screen (the axes of native-describe-screen frames); on a Mac app, points from the top-left of the area describe reports. NOT normalized [0,1] tap space. */
+        /** Raw X coordinate in points, on the portrait axes of the screen; on a Mac app, points from the top-left of the area describe reports. NOT normalized [0,1] tap space. */
         x: number;
         /** Raw Y coordinate in points, on the same axes as x. NOT normalized [0,1] tap space. */
         y: number;
@@ -590,7 +590,7 @@ export interface ArgentToolArgs {
         includeAncestors?: boolean;
         /** Include child views for the matched view (default false) */
         includeChildren?: boolean;
-        /** View fields to include. Defaults: pointer, className, tag, frame, windowFrame, screenFrame, bounds, hidden, alpha, opaque, clipsToBounds, userInteractionEnabled, depth, identifier, label, layerName, nativeID. Additional: center, transform, contentMode, backgroundColor, tintColor */
+        /** View fields to include. Defaults: pointer, className, tag, frame, windowFrame, screenFrame, bounds, hidden, alpha, opaque, clipsToBounds, userInteractionEnabled, depth, identifier, label, layerName, nativeID. Additional: center, transform, contentMode, backgroundColor, tintColor, value, firstResponder */
         fields?: Array<string>;
         /** Exact UIView class names whose entire subtree should be pruned */
         skipClasses?: Array<string>;

package/dist/tool-server.cjs

@@ -101252,7 +101252,7 @@ var _CI_VENDOR_COUNT_FOR_TEST = vendors_default.length;
 var SESSION_ID = (0, import_node_crypto3.randomUUID)();
 function readCliVersion() {
   if (true) {
-    return "0.32.1-next.17";
+    return "0.32.1-next.18";
   }
   return "0.0.0";
 }
@@ -123698,7 +123698,7 @@ var NATIVE_DEVTOOLS_NAMESPACE = "NativeDevtools";
 function isInjectableBundleId(bundleId) {
   return !bundleId.toLowerCase().startsWith("com.apple.");
 }
-var NON_INJECTABLE_NATIVE_WARNING = "Do not fall back to the native-devtools feature tools (native-describe-screen, native-find-views, native-full-hierarchy, native-network-logs, native-view-at-point, native-user-interactable-view-at-point): they run the same injection precheck and fail with the same non-injectable error.";
+var NON_INJECTABLE_NATIVE_WARNING = "Do not fall back to the native-devtools feature tools (native-find-views, native-full-hierarchy, native-network-logs, native-view-at-point, native-user-interactable-view-at-point): they run the same injection precheck and fail with the same non-injectable error.";
 var NON_INJECTABLE_RECOVERY = "Use the standard `describe` tool (its accessibility path reads the screen without injection) or `screenshot` (then interact by coordinate). " + NON_INJECTABLE_NATIVE_WARNING;
 var MAX_NATIVE_DEVTOOLS_INIT_ATTEMPTS = 3;
 var NATIVE_DEVTOOLS_AGE_SLOP_MS = 3e3;
@@ -123865,7 +123865,7 @@ var nativeDevtoolsBlueprint = {
     const { device } = opts;
     if (device.platform !== "ios" && device.platform !== "ios-remote") {
       throw new FailureError(
-        `${NATIVE_DEVTOOLS_NAMESPACE} is iOS-only. The target '${device.id}' classifies as ${device.platform} \u2014 native-devtools tools (native-describe-screen, native-find-views, etc.) only drive iOS simulators. Pick an iOS udid from list-devices.`,
+        `${NATIVE_DEVTOOLS_NAMESPACE} is iOS-only. The target '${device.id}' classifies as ${device.platform} \u2014 native-devtools tools (native-find-views, native-view-at-point, etc.) only drive iOS simulators. Pick an iOS udid from list-devices.`,
         {
           error_code: FAILURE_CODES.NATIVE_DEVTOOLS_WRONG_PLATFORM,
           failure_stage: "native_devtools_factory_options",
@@ -128234,6 +128234,8 @@ async function listRemoteIosSimulators() {
     const result = await simctlListDevices();
     const out = [];
     for (const [runtime, devices] of Object.entries(result.devices)) {
+      const runtimeKind = simulatorRuntimeKind(runtime);
+      if (!runtimeKind) continue;
       for (const d of devices) {
         if (d.isAvailable === false) continue;
         out.push({
@@ -128242,7 +128244,7 @@ async function listRemoteIosSimulators() {
           name: d.name,
           state: d.state,
           runtime,
-          runtimeKind: simulatorRuntimeKind(runtime)
+          runtimeKind
         });
       }
     }
@@ -129112,6 +129114,7 @@ var describeNodeSchema = external_exports.lazy(
     frame: describeFrameSchema,
     children: external_exports.array(describeNodeSchema),
     label: external_exports.string().optional(),
+    hint: external_exports.string().optional(),
     identifier: external_exports.string().optional(),
     value: external_exports.string().optional(),
     subtreeText: external_exports.string().optional(),
@@ -129191,8 +129194,11 @@ function adaptNativeDescribeElementToDescribeNode(element) {
     frame: frame2,
     children: [],
     label: element.label,
+    hint: element.hint,
     identifier: element.identifier,
-    value: element.value
+    value: element.value,
+    ...element.traits.includes("notEnabled") ? { disabled: true } : {},
+    ...element.traits.includes("selected") ? { selected: true } : {}
   };
 }
 function adaptNativeDescribeToDescribeResult(result) {
@@ -129236,10 +129242,13 @@ function adaptAXElement(el) {
     },
     children: [],
     label: el.label,
+    hint: el.hint,
     value: el.value,
     identifier: el.identifier,
     // The text input that has the keyboard.
-    ...traits.includes("isEditing") ? { focused: true } : {}
+    ...traits.includes("isEditing") ? { focused: true } : {},
+    ...traits.includes("notEnabled") ? { disabled: true } : {},
+    ...traits.includes("selected") ? { selected: true } : {}
   };
 }
 function adaptAXDescribeToDescribeResult(response) {
@@ -146136,7 +146145,7 @@ Returns { envSetup, appRunning, connected, requiresRestart, state, message, next
 - nextLaunchWillBeInjected: if you launch this bundle now, native devtools env setup is already in place (always false for a non-injectable app)
 - injectable: false for Apple system apps (bundle ids under com.apple.), which the native tools refuse to read.
 
-If injectable is false: treat this as TERMINAL. The app is not a supported native-devtools target, and no relaunch changes that. Do NOT restart/retry. Use the standard \`describe\` tool (its accessibility path reads the screen without injection) or \`screenshot\` (then interact by coordinate). Do not fall back to the native-devtools feature tools (native-describe-screen, native-find-views, native-full-hierarchy, native-network-logs, native-view-at-point, native-user-interactable-view-at-point): they run the same injection precheck and fail with the same non-injectable error.
+If injectable is false: treat this as TERMINAL. The app is not a supported native-devtools target, and no relaunch changes that. Do NOT restart/retry. Use the standard \`describe\` tool (its accessibility path reads the screen without injection) or \`screenshot\` (then interact by coordinate). Do not fall back to the native-devtools feature tools (native-find-views, native-full-hierarchy, native-network-logs, native-view-at-point, native-user-interactable-view-at-point): they run the same injection precheck and fail with the same non-injectable error.
 If appRunning is false and nextLaunchWillBeInjected is true: use launch-app normally.
 If requiresRestart is true: call restart-app once, then proceed with the native feature. Read state before acting on a second such reading. Indeterminate reaches this rule too, and its line below bounds it at that one restart.
 If state is unregistered: do NOT restart the app again: it already launched under the terms a restart would recreate. Restart the tool-server (\`argent server stop && argent server start --detach\`), then retry. If it reads unregistered again after that restart, stop: the process loads argent's dylib but never dials, and no further restart on either side changes it. Treat native devtools as unavailable, then use \`describe\` or \`screenshot\` and drive by coordinate.
@@ -146508,15 +146517,15 @@ var zodSchema5 = external_exports.object({
   includeAncestors: external_exports.boolean().optional().describe("Include ancestor chain for each matched view (default true)"),
   includeChildren: external_exports.boolean().optional().describe("Include child views for each matched view (default true)"),
   fields: external_exports.array(external_exports.string()).optional().describe(
-    "View fields to include. Defaults: className, frame, hidden, alpha, identifier, label, nativeID, userInteractionEnabled, depth. Additional: pointer, tag, windowFrame, screenFrame, bounds, center, opaque, clipsToBounds, transform, contentMode, backgroundColor, tintColor, layerName"
+    "View fields to include. Defaults: pointer, className, tag, frame, windowFrame, screenFrame, bounds, hidden, alpha, opaque, clipsToBounds, userInteractionEnabled, depth, identifier, label, layerName, nativeID. Additional: center, transform, contentMode, backgroundColor, tintColor, value, firstResponder"
   )
 });
 var nativeFindViewsTool = {
   id: "native-find-views",
   description: `Search for specific UIViews in the running app by class name, accessibility identifier, accessibility label, tag or React Native nativeID; at least one of these params is required.
 Use when you need to locate a specific view by its properties without dumping the entire native-full-hierarchy.
-Returns { status: "ok", matches }: the matching views with their frames, properties, optional ancestors and optional children.
-<flag:macos-apps>On a Mac app it searches the NSViews of every visible window: className is the NSView class, and identifier matches the accessibility identifier or the view identifier. The statuses below are iOS only.
+Returns { status: "ok", matches }: the matching views with their frames, properties, optional ancestors and optional children. Frames are in points. frame and center are relative to the superview, and windowFrame turns with the UI. To tap a view, divide the centre of its screenFrame by the screen size in points, which native-full-hierarchy returns as screen (maxDepth: 1 is enough), and pass the result to gesture-tap. Matches are not filtered for visibility. Before tapping, call native-user-interactable-view-at-point at that centre with skipClasses ["UITextEffectsWindow"]. If it does not return the match or a descendant, check a screenshot.
+<flag:macos-apps>On a Mac app it searches the NSViews of every visible window: className is the NSView class, and identifier matches the accessibility identifier or the view identifier. The tap recipe above and the statuses below are iOS only.
 </flag:macos-apps>If status is restart_required: follow the message (usually restart-app), then retry. If status is service_stale: the app is already injected, so restarting it cannot help. Restart the tool-server (\`argent server stop && argent server start --detach\`) and retry. If the same status comes back after that restart, stop restarting: follow the message, which names the terminal fallback. If status is connect_pending: the app is injected and still connecting. Do not restart it, wait a few seconds and retry. If status is init_failed: follow the message (re-boot the simulator) rather than retrying.
 An Apple system app is rejected outright (terminal: never retry it).`,
   zodSchema: zodSchema5,
@@ -146553,7 +146562,7 @@ init_zod();
 var zodSchema6 = external_exports.object({
   ...nativeAppTarget,
   fields: external_exports.array(external_exports.string()).optional().describe(
-    "View fields to include. Use EXACT names: className, frame, hidden, alpha, identifier, label, nativeID, userInteractionEnabled, depth, pointer, tag, windowFrame, screenFrame, bounds, center, opaque, clipsToBounds, transform, contentMode, backgroundColor, tintColor, layerName. Defaults to all of the first group when omitted."
+    "View fields to include. Use EXACT names. Defaults: pointer, className, tag, frame, windowFrame, screenFrame, bounds, hidden, alpha, opaque, clipsToBounds, userInteractionEnabled, depth, identifier, label, layerName, nativeID. Additional: center, transform, contentMode, backgroundColor, tintColor, value, firstResponder"
   ),
   skipClasses: external_exports.array(external_exports.string()).optional().describe(
     'Exact UIView class names whose entire subtree should be pruned (e.g. ["UIImageView"] to drop image leaf nodes)'
@@ -146636,6 +146645,10 @@ normalized contract.
 An Apple system app is rejected outright (terminal: never retry it).`,
   zodSchema: zodSchema7,
   readOnly: true,
+  // Agents get the disabled and selected states and the VoiceOver hint from
+  // `describe`, which falls back to this RPC (ViewHierarchy.describeScreen) when
+  // the ax-service returns nothing. The SDK still calls this tool directly.
+  hideFromMcp: true,
   interaction: {
     startedMsg: ({ params }) => `Reading native screen for ${params.bundleId}`,
     completedMsg: ({ params }) => `Read native screen for ${params.bundleId}`,
@@ -146663,13 +146676,13 @@ init_src2();
 var zodSchema8 = external_exports.object({
   ...nativeAppTarget,
   x: external_exports.number().describe(
-    "Raw X coordinate in points, on the portrait axes of the screen (the axes of native-describe-screen frames)<flag:macos-apps>; on a Mac app, points from the top-left of the area describe reports</flag:macos-apps>. NOT normalized [0,1] tap space."
+    "Raw X coordinate in points, on the portrait axes of the screen<flag:macos-apps>; on a Mac app, points from the top-left of the area describe reports</flag:macos-apps>. NOT normalized [0,1] tap space."
   ),
   y: external_exports.number().describe("Raw Y coordinate in points, on the same axes as x. NOT normalized [0,1] tap space."),
   includeAncestors: external_exports.boolean().optional().describe("Include ancestor chain for the matched view (default true)"),
   includeChildren: external_exports.boolean().optional().describe("Include child views for the matched view (default false)"),
   fields: external_exports.array(external_exports.string()).optional().describe(
-    "View fields to include. Defaults: pointer, className, tag, frame, windowFrame, screenFrame, bounds, hidden, alpha, opaque, clipsToBounds, userInteractionEnabled, depth, identifier, label, layerName, nativeID. Additional: center, transform, contentMode, backgroundColor, tintColor"
+    "View fields to include. Defaults: pointer, className, tag, frame, windowFrame, screenFrame, bounds, hidden, alpha, opaque, clipsToBounds, userInteractionEnabled, depth, identifier, label, layerName, nativeID. Additional: center, transform, contentMode, backgroundColor, tintColor, value, firstResponder"
   ),
   skipClasses: external_exports.array(external_exports.string()).optional().describe("Exact UIView class names whose entire subtree should be pruned"),
   skipClassPrefixes: external_exports.array(external_exports.string()).optional().describe("Class name prefixes to prune entire subtrees"),
@@ -146691,7 +146704,8 @@ identifier, label and layer name, plus its ancestors and subviews as requested.
 view is null when nothing is drawn at that point.
 
 To get x and y from a normalized describe frame, multiply its values by the
-screen size in points, also when the UI is landscape.<flag:macos-apps> On a Mac app,
+screen size in points, also when the UI is landscape.
+native-full-hierarchy returns that size as screen (pass maxDepth: 1 and fields: ["className"] to keep it small).<flag:macos-apps> On a Mac app,
 multiply them by the size of the area describe reports instead.</flag:macos-apps>
 
 If status is restart_required: follow the message (usually restart-app), then retry. If status is service_stale: the app is already injected, so restarting it cannot help. Restart the tool-server (\`argent server stop && argent server start --detach\`) and retry. If the same status comes back after that restart, stop restarting: follow the message, which names the terminal fallback. If status is connect_pending: the app is injected and still connecting. Do not restart it, wait a few seconds and retry. If status is init_failed: follow the message (re-boot the simulator) rather than retrying.
@@ -146725,13 +146739,13 @@ init_src2();
 var zodSchema9 = external_exports.object({
   ...nativeAppTarget,
   x: external_exports.number().describe(
-    "Raw X coordinate in points, on the portrait axes of the screen (the axes of native-describe-screen frames)<flag:macos-apps>; on a Mac app, points from the top-left of the area describe reports</flag:macos-apps>. NOT normalized [0,1] tap space."
+    "Raw X coordinate in points, on the portrait axes of the screen<flag:macos-apps>; on a Mac app, points from the top-left of the area describe reports</flag:macos-apps>. NOT normalized [0,1] tap space."
   ),
   y: external_exports.number().describe("Raw Y coordinate in points, on the same axes as x. NOT normalized [0,1] tap space."),
   includeAncestors: external_exports.boolean().optional().describe("Include ancestor chain for the matched view (default true)"),
   includeChildren: external_exports.boolean().optional().describe("Include child views for the matched view (default false)"),
   fields: external_exports.array(external_exports.string()).optional().describe(
-    "View fields to include. Defaults: pointer, className, tag, frame, windowFrame, screenFrame, bounds, hidden, alpha, opaque, clipsToBounds, userInteractionEnabled, depth, identifier, label, layerName, nativeID. Additional: center, transform, contentMode, backgroundColor, tintColor"
+    "View fields to include. Defaults: pointer, className, tag, frame, windowFrame, screenFrame, bounds, hidden, alpha, opaque, clipsToBounds, userInteractionEnabled, depth, identifier, label, layerName, nativeID. Additional: center, transform, contentMode, backgroundColor, tintColor, value, firstResponder"
   ),
   skipClasses: external_exports.array(external_exports.string()).optional().describe("Exact UIView class names whose entire subtree should be pruned"),
   skipClassPrefixes: external_exports.array(external_exports.string()).optional().describe("Class name prefixes to prune entire subtrees"),
@@ -146753,7 +146767,8 @@ identifier, label and layer name, plus its ancestors and subviews as requested.
 view is null when no touchable view sits under that point.
 
 To get x and y from a normalized describe frame, multiply its values by the
-screen size in points, also when the UI is landscape.<flag:macos-apps> On a Mac app,
+screen size in points, also when the UI is landscape.
+native-full-hierarchy returns that size as screen (pass maxDepth: 1 and fields: ["className"] to keep it small).<flag:macos-apps> On a Mac app,
 multiply them by the size of the area describe reports instead.</flag:macos-apps>
 
 If status is restart_required: follow the message (usually restart-app), then retry. If status is service_stale: the app is already injected, so restarting it cannot help. Restart the tool-server (\`argent server stop && argent server start --detach\`) and retry. If the same status comes back after that restart, stop restarting: follow the message, which names the terminal fallback. If status is connect_pending: the app is injected and still connecting. Do not restart it, wait a few seconds and retry. If status is init_failed: follow the message (re-boot the simulator) rather than retrying.
@@ -154084,8 +154099,7 @@ var gestureTapTool = {
   description: `Press the device screen (iOS simulator or physical device, Android emulator, <flag:harmonyos>HarmonyOS device, </flag:harmonyos><flag:macos-apps>Mac app, </flag:macos-apps>or Chromium app) at a point given as fractions of the screen width and height (not pixels): a Down event followed by an Up event at the same spot.<flag:macos-apps> On a Mac app it is a click, and x/y are fractions of the area describe reports, not of the window.</flag:macos-apps>
 Use when tapping a button, link, or any tappable element on the screen.
 Returns { tapped: true, timestampMs }. On physical iOS, reactivated: true = app was re-fronted; re-describe. Fails if the device backend (simulator-server / emulator / Chromium CDP<flag:harmonyos> / \`hdc\`</flag:harmonyos>) is unreachable.
-On a physical iPhone use \`describe\`; \`native-describe-screen\` is simulator-only.
-Before tapping, determine the correct coordinates by using discovery tools. Pick by platform: iOS / Android use \`describe\` (on an iOS simulator, also \`native-describe-screen\`); Chromium uses \`describe\` (the DOM walker)<flag:harmonyos> and HarmonyOS uses \`describe\` (the \`uitest\` layout dump)</flag:harmonyos>, since the native discovery tools don't apply<flag:harmonyos> to either</flag:harmonyos><flag:macos-apps>; a Mac app uses \`describe\`</flag:macos-apps>. More information in \`argent-device-interact\` skill`,
+Before tapping, take the coordinates from \`describe\` on every platform (on Chromium it walks the DOM<flag:harmonyos>, on HarmonyOS it reads the \`uitest\` layout dump</flag:harmonyos>). More information in \`argent-device-interact\` skill`,
   zodSchema: zodSchema18,
   alwaysLoad: true,
   // A physical iPhone's first call waits out the runner build, past the MCP fetch timeout,
@@ -160089,19 +160103,24 @@ function hasContent(n) {
 function shouldEmit(n, contentRoles) {
   return hasContent(n) || contentRoles.has(n.role);
 }
+function hasIosState(n) {
+  return Boolean(n.hint || n.disabled || n.selected);
+}
 function formatLine(n, indent, source) {
   const pad = "  ".repeat(indent);
   const dedupedValue = n.value && n.value !== n.label ? n.value : void 0;
+  const dedupedHint = n.hint && n.hint !== n.label ? n.hint : void 0;
   const labelPart = formatLabel(n.label);
   const valuePart = formatAttr("value", dedupedValue);
   const idPart = formatAttr("id", n.identifier);
+  const hintPart = formatAttr("hint", dedupedHint);
   const flagPart = formatFlags(n, source);
-  const annotations = `${labelPart}${valuePart}${idPart}${flagPart}`.trim();
+  const annotations = `${labelPart}${valuePart}${idPart}${hintPart}${flagPart}`.trim();
   const annotated = annotations ? ` ${annotations}` : "";
   return `${pad}${n.role}${annotated}  ${fmtFrame(n.frame)}`;
 }
 function renderFlat(root, contentRoles, source) {
-  return root.children.filter((n) => shouldEmit(n, contentRoles)).slice().sort((a, b) => a.frame.y - b.frame.y || a.frame.x - b.frame.x).map((n) => formatLine(n, 1, source));
+  return root.children.filter((n) => shouldEmit(n, contentRoles) || hasIosState(n)).slice().sort((a, b) => a.frame.y - b.frame.y || a.frame.x - b.frame.x).map((n) => formatLine(n, 1, source));
 }
 function renderNested(root, contentRoles, source) {
   const lines = [];
@@ -160317,11 +160336,10 @@ On HarmonyOS it runs the device's \`uitest dumpLayout\` (\`source: "harmony-uite
 On a Mac app it reads the app's accessibility tree with web view content and React Native testIDs merged in; frames are fractions of the app window, grown to include sheets and menus past it.</flag:macos-apps>
 On a TV target (Apple TV / Android TV: a \`list-devices\` entry with runtimeKind 'tv') it returns the FOCUSED element and the FOCUSABLE elements instead, since a TV UI has no tap coordinates. Move the highlight with \`tv-remote\`, then call describe again to confirm where focus landed.
 
-Returns \`{ description, source }\`: \`description\` renders the tree one line per element, with its role, label/value/id, interactivity flags and frame. Frames are normalized [0,1] fractions of the screen / window width and height (not pixels); tap an element at the centre of its frame. The tree has no z-order or occlusion information: an element may be covered by an overlay (e.g. a toolbar over list rows), so check a screenshot when a tap lands unexpectedly.
+Returns \`{ description, source }\`: \`description\` renders the tree one line per element, with its role, label/value/id, VoiceOver hint (iOS simulator), interactivity flags and frame. Frames are normalized [0,1] fractions of the screen / window width and height (not pixels); tap an element at the centre of its frame. The tree has no z-order or occlusion information: an element may be covered by an overlay (e.g. a toolbar over list rows), so check a screenshot when a tap lands unexpectedly.
 Fails if the tree cannot be read, e.g. on a physical iOS device before launch-app has started an app.
 
-For app-scoped inspection with full UIKit properties (accessibilityIdentifier, viewClassName),
-use native-describe-screen with an explicit bundleId instead (iOS simulator only).
+For UIKit properties of a view (class name, accessibilityIdentifier), use native-find-views (iOS simulator only).
 For React Native component names, or a testID describe does not list, use debugger-component-tree.`,
     zodSchema: zodSchema40,
     alwaysLoad: true,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@swmansion/argent",
-  "version": "0.32.1-next.17",
+  "version": "0.32.1-next.19",
   "mcpName": "io.github.software-mansion/argent",
   "description": "MCP server for iOS Simulator, Android Emulator, Chromium app and Mac app control",
   "license": "Apache-2.0",

package/skills/argent-create-e2e/references/reliability-and-recovery.md

@@ -26,7 +26,7 @@ Convert element-seeking swipes to `scroll-to`. Keep a coordinate swipe only when
 
 Work this gate as soon as capture warns that it kept a raw point or recorded a role-only selector. Keep the source screen available and do these checks:
 
-1. **iOS:** query plausible ids or labels with `native-find-views`. If no term is useful, call `native-full-hierarchy` with narrow fields and `maxDepth: 100`. `describe` and `native-describe-screen` are accessibility projections. They cannot prove that no YAML selector exists.
+1. **iOS:** query plausible ids or labels with `native-find-views`. If no term is useful, call `native-full-hierarchy` with narrow fields and `maxDepth: 100`. `describe` is an accessibility projection. It cannot prove that no YAML selector exists.
 2. **Other platforms:** use `describe`; for React Native on Android, also `debugger-component-tree`, for testIDs `describe` omits. Verify Android and Chromium candidates in step 3. Their discovery trees can omit runner elements.
 3. Test each candidate in a scratch fragment with `assert: { visible: <candidate> }` on the valid screen. Inspect every failure before trying a better id, label, app, or container.
 4. If source is available, inspect its `testID`, `accessibilityIdentifier`, or `resource-id`. If none exists, report the missing stable id as the real fix.

package/skills/argent-device-interact/SKILL.md

@@ -45,7 +45,7 @@ After an action moves you to a different screen, or whenever you don't know an e
 | Need                                           | Tool                                                                                                                           |
 | ---------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------ |
 | Any screen: app, system dialog, Home screen    | `describe`                                                                                                                     |
-| App-scoped UIKit properties (iOS simulator)    | `native-describe-screen` (needs `bundleId`)                                                                                    |
+| UIKit properties of a view (iOS simulator)     | `native-find-views` (needs `bundleId` and a class, id, label, tag or `nativeID` to match)                                      |
 | What sits at a known raw point (iOS simulator) | `native-user-interactable-view-at-point` (the view that receives touch) or `native-view-at-point` (the visually deepest view)  |
 | Final visual fallback                          | `screenshot`, only when the tools above cannot expose the control. Do not derive routine in-app navigation targets from pixels |
 

package/skills/argent-device-interact/references/harmonyos.md

@@ -17,7 +17,7 @@ Refused as unsupported on harmony:
 
 - `gesture-rotate` and `gesture-custom`: the device moves each contact along one straight line per call, so there is no arc and no long-press. There is no long-press tool on HarmonyOS.
 - `gesture-drag` and `gesture-scroll` (Chromium-only), and `rotate`: the device has no orientation command.
-- `native-describe-screen`, `native-view-at-point`, `native-user-interactable-view-at-point`, `debugger-component-tree`, the rest of the debugger (`debugger-evaluate`, `debugger-log-registry`), `view-network-logs` and the profilers.
+- `native-view-at-point`, `native-user-interactable-view-at-point`, `debugger-component-tree`, the rest of the debugger (`debugger-evaluate`, `debugger-log-registry`), `view-network-logs` and the profilers.
 - `paste`, so a `{{secret:...}}` placeholder goes through `keyboard` only.
 - Screen recording: HarmonyOS exposes no readable frame stream. Use `screenshot` for a still frame.
 

package/skills/argent-ios-simulator-setup/SKILL.md

@@ -5,7 +5,7 @@ description: Set up and connect to an iOS simulator using argent MCP tools. Use
 
 ## 1. Setup Steps
 
-1. **Find a booted simulator.** Use `list-devices`. Filter for entries with `platform: "ios"` and skip `kind: "device"` (a physical iPhone): booted simulators are listed first. If no fitting simulator is booted, call `boot-device` with `udid: <chosen UDID>`.
+1. **Find a booted simulator.** Use `list-devices`. Filter for entries with `platform: "ios"` and skip `kind: "device"` (a physical iPhone): booted simulators are listed first. Through MCP, simulators that are shut down follow the JSON object, one line each, with the udid second. If no fitting simulator is booted, call `boot-device` with `udid: <chosen UDID>`.
 2. **No connect step.** Interaction tools (`gesture-tap`, `describe`, `screenshot`, …) auto-start the simulator server on first use.
 
 ## 2. Notes

package/skills/argent-screenshot-diff/SKILL.md

@@ -7,7 +7,7 @@ description: Compare saved or live app screenshots with the argent screenshot-di
 
 Use `screenshot-diff` as supporting visual evidence for UI QA and visual regression checks. It highlights pixel-visible change or stability; it does not replace visual inspection, accessibility/component-tree state, frame/attribute checks, logs, network evidence, or app behavior. Its description covers the input combinations, physical-iPhone captures and the result.
 
-Do not use screenshot diffing for tap-coordinate discovery. Use `describe` or `native-describe-screen` to find targets first.
+Do not use screenshot diffing for tap-coordinate discovery. Use `describe` to find targets first.
 
 ## 2. When To Use
 

package/skills/argent-test-ui-flow/SKILL.md

@@ -20,7 +20,7 @@ For implementation tasks that modify visible UI, this workflow also serves as a
 3. **Interact**: perform the action (`gesture-tap`, `gesture-swipe`, `keyboard`, `button`, ...); you receive a screenshot automatically.
 4. **Verify**: check the returned screenshot. If it shows a loading/transitional state, block with `await-ui-element` (expected element `visible`, or a spinner `hidden`) rather than a guessed delay (its result carries the new screenshot and element tree), but only with a selector you can trust (`text`/`identifier`/`role`) that the screen is known to have or that you saw in a prior `describe`; a guessed one just times out. Otherwise use a short fixed wait. Pick evidence by what's being asserted:
    - **Visual** (layout, spacing, color, typography, image/icon rendering, clipping, overflow, text rendering): prefer `screenshot-diff` against the step 1 baseline: it surfaces pixel-visible changes the auto-screenshot might miss. Fall back to inspecting the auto-screenshot only when a stable baseline isn't available.
-   - **Structural** (navigation state, element existence, accessibility labels/values, selection, hierarchy, route): verify with `describe`, `debugger-component-tree`, or `native-describe-screen`.
+   - **Structural** (navigation state, element existence, accessibility labels/values, selection, hierarchy, route): verify with `describe` or `debugger-component-tree`.
    - **Runtime / log / network** (console errors, API calls, persistence, timing): verify with `view-network-logs`, `debugger-log-registry`, `debugger-evaluate`, or targeted tests. A `not_connected` result from the debugger tools is not evidence about the app: follow its `guidance`, then re-verify.
    - **Mixed**: collect evidence for each relevant class.
    - Report the combined verdict: expected behavior, observed behavior, evidence used, and any blocker for requested visual diffing.