@momasiku/dsh-pilot 0.5.8 → 0.5.10

package/README.md

@@ -4,10 +4,12 @@
 
 **Give DeepSeek Harness hands and eyes on your desktop.**
 
+[![npm](https://img.shields.io/npm/v/@momasiku/dsh-pilot?label=npm)](https://www.npmjs.com/package/@momasiku/dsh-pilot)
+[![compat](https://github.com/momasiku/dsh-pilot/actions/workflows/compat.yml/badge.svg)](https://github.com/momasiku/dsh-pilot/actions/workflows/compat.yml)
 [![license](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
+
 ![platform](https://img.shields.io/badge/platform-Windows-0078d4.svg)
-[![npm](https://img.shields.io/npm/v/@momasiku/dsh-pilot?label=npm)](https://www.npmjs.com/package/@momasiku/dsh-pilot)
-![tools](https://img.shields.io/badge/tools-screen__view%20%C2%B7%20desktop__control%20%C2%B7%20desktop__sequence-success)
+![DSH](https://img.shields.io/badge/DSH-0.1.5--rc.2%20%E2%86%92%200.2.x-5b50ed.svg)
 
 [English](#why-this-exists) · [中文说明](#中文说明)
 

package/lib/index.js

@@ -992,7 +992,7 @@ const POINTER_ACTIONS = new Set(['click', 'doubleClick', 'rightClick', 'middleCl
 const RISKS = ['low', 'medium', 'high'];
 const RISK_ORDER = { low: 0, medium: 1, high: 2 };
 /** Step fields forwarded verbatim to the action helper. */
-const STEP_FIELDS = ['x', 'y', 'toX', 'toY', 'amount', 'text', 'key', 'title', 'button'];
+const STEP_FIELDS = ['x', 'y', 'toX', 'toY', 'amount', 'text', 'key', 'title', 'button', 'tolerant'];
 
 const SEQUENCE_STEP_RESULT_SCHEMA = {
 	type: 'object',
@@ -1281,13 +1281,17 @@ function apply(ctx, config) {
 				type: 'string',
 				description: 'Case-insensitive window title or process substring; required by focus, and filters windows.'
 			},
+			tolerant: {
+				type: 'boolean',
+				description: 'focus only: when Windows refuses to hand the window the foreground, record a note and continue instead of failing the step. Use it when the coordinates are right whichever window ends up in front.'
+			},
 			capture: {
 				type: 'boolean',
 				description: 'Capture a fresh frame after the action and show it in this result. Default true.'
 			},
 			settleMs: {
 				type: 'integer',
-				description: 'Milliseconds to wait before the automatic capture, so the screen can repaint. Default 750.'
+				description: 'Milliseconds to wait before the automatic capture, so the screen can repaint. Default 750. Skipped when capture is false, unless you pass it explicitly.'
 			},
 			forceImage: {
 				type: 'boolean',
@@ -1338,9 +1342,15 @@ function apply(ctx, config) {
 			const wantsCapture = args.capture !== false;
 			if (wantsCapture) await assertImageCapableRoute(ctx, exec, 'desktop_control with capture');
 
-			const settle = Number.isSafeInteger(args.settleMs) ? args.settleMs : config.settleMs;
+			// The settle exists so a frame is not taken mid-repaint. With no frame to
+			// take there is nothing to wait for, and the wait is dead time: measured
+			// 750 ms on every desktop_control call, capturing or not. An explicit
+			// settleMs is still honoured, so a caller that wants a pause before its
+			// own screen_view keeps it.
+			const explicitSettle = Number.isSafeInteger(args.settleMs);
+			const settle = explicitSettle ? args.settleMs : (wantsCapture ? config.settleMs : 0);
 			const params = { action: args.action, settleMs: settle };
-			for (const field of ['x', 'y', 'toX', 'toY', 'amount', 'text', 'key', 'title', 'button']) {
+			for (const field of ['x', 'y', 'toX', 'toY', 'amount', 'text', 'key', 'title', 'button', 'tolerant']) {
 				if (args[field] !== undefined && args[field] !== null) params[field] = args[field];
 			}
 
@@ -1396,7 +1406,7 @@ function apply(ctx, config) {
 			steps: {
 				type: 'array',
 				required: true,
-				description: 'Ordered steps. Each object is one desktop_control action: {action, x, y, toX, toY, text, key, title, amount, button, settleMs, risk, riskNote}; action "wait" takes ms. Coordinates are physical screen pixels.'
+				description: 'Ordered steps. Each object is one desktop_control action: {action, x, y, toX, toY, text, key, title, amount, button, tolerant, settleMs, risk, riskNote}; action "wait" takes ms. A focus step that Windows refuses now fails unless tolerant: true. Coordinates are physical screen pixels.'
 			},
 			risk: {
 				type: 'string',

package/lib/scripts/_dsh-action.ps1

@@ -304,21 +304,31 @@ switch ($action) {
         }
     }
     'focus' {
+        # A refused hand-off used to be a note: the step reported ok and the batch
+        # carried on, so every later click and keystroke went to whatever window was
+        # actually in front. That is the one failure this plugin must not report as
+        # success, so it is an error now; tolerant keeps the old behaviour for a
+        # caller that knows the coordinates are right either way.
+        $tolerant = $false
+        $tolerantRaw = Get-Param 'tolerant'
+        if ($null -ne $tolerantRaw) { $tolerant = [bool] $tolerantRaw }
+        $label = ''
+        $target = [IntPtr]::Zero
         if ($script:TargetHwnd -ne [IntPtr]::Zero) {
             # An explicit handle is authoritative and works for a window hidden
             # to the tray, which a title search cannot reach.
             if (-not [DshInput]::IsAlive($script:TargetHwnd)) { Write-ActionError "the target window (hwnd $($script:TargetHwndText)) no longer exists" }
-            $raised = [DshInput]::ForceForeground($script:TargetHwnd)
-            if (-not $raised) { $notes += "Windows would not hand focus to hwnd $($script:TargetHwndText); '$([DshInput]::ForegroundTitle())' kept it." }
+            $target = $script:TargetHwnd
+            $label = "hwnd $($script:TargetHwndText)"
+            $null = [DshInput]::ForceForeground($target)
             Start-Sleep -Milliseconds 220
         } else {
             if ([string]::IsNullOrWhiteSpace($windowTitle)) { Write-ActionError 'focus requires a title, or an hwnd' }
             $needle = $windowTitle.ToLowerInvariant()
             # EnumWindows, not Process.MainWindowHandle: an Electron, UWP or tray
             # window frequently reports no main window handle at all, so a
-            # MainWindowHandle scan cannot find a window that `windows` (which
-            # already enumerates) happily lists.
-            $target = [IntPtr]::Zero
+            # MainWindowHandle scan cannot find a window that the windows action
+            # (which already enumerates) happily lists.
             $fallback = [IntPtr]::Zero
             foreach ($handle in [DshInput]::TopLevelWindows()) {
                 if (-not [DshInput]::IsWindowVisible($handle)) { continue }
@@ -335,10 +345,22 @@ switch ($action) {
             }
             if ($target -eq [IntPtr]::Zero) { $target = $fallback }
             if ($target -eq [IntPtr]::Zero) { Write-ActionError "no visible top-level window matches '$windowTitle'; pass an hwnd if the window is hidden to the tray" }
-            $raised = [DshInput]::Activate($target)
-            if (-not $raised) { $notes += "Windows refused the foreground request for '$([DshInput]::WindowTitle($target))'; the taskbar entry may still need a click." }
+            $label = "'" + [DshInput]::WindowTitle($target) + "'"
+            $null = [DshInput]::Activate($target)
             Start-Sleep -Milliseconds 220
         }
+
+        # Ask the desktop, not the API. SetForegroundWindow is documented to be
+        # unreliable about its return value, and the state that decides where the
+        # next click lands is the foreground window, not what the call returned.
+        if (-not [DshInput]::IsForeground($target)) {
+            $holder = [DshInput]::ForegroundTitle()
+            if ($tolerant) {
+                $notes += "Windows would not hand focus to $label; '$holder' kept it (tolerant: continuing)."
+            } else {
+                Write-ActionError "could not bring $label to the foreground; '$holder' still holds it, so the rest of this batch would land there. Pass tolerant: true on this step if the coordinates are right either way."
+            }
+        }
     }
     'windows' {
         $query = $windowTitle.ToLowerInvariant()

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@momasiku/dsh-pilot",
-  "version": "0.5.8",
+  "version": "0.5.10",
   "description": "Desktop vision and control for DeepSeek Harness - the whole Windows machine, any application, not browser automation: screen_view captures the live screen as an image the model can see, desktop_control drives the mouse, keyboard, and windows against the exact same pixel space; desktop_sequence runs a whole batch of actions in one call and looks once at the end.",
   "license": "MIT",
   "author": "momasiku",
@@ -40,9 +40,9 @@
   },
   "peerDependencies": {
     "@deepseek-ai/cordis": "^4.0.2",
-    "@deepseek-ai/dsh-attachment": ">=0.1.5-rc.2 <0.2.0 || >=0.2.0-rc.1 <0.3.0-0",
-    "@deepseek-ai/dsh-llm": ">=0.1.5-rc.2 <0.2.0 || >=0.2.0-rc.1 <0.3.0-0",
-    "@deepseek-ai/dsh-tools": ">=0.1.5-rc.2 <0.2.0 || >=0.2.0-rc.1 <0.3.0-0"
+    "@deepseek-ai/dsh-attachment": ">=0.1.5-rc.2 <0.2.0 || >=0.2.0-rc.1 <0.3.0-0 || >=0.2.1-0 <0.3.0-0",
+    "@deepseek-ai/dsh-llm": ">=0.1.5-rc.2 <0.2.0 || >=0.2.0-rc.1 <0.3.0-0 || >=0.2.1-0 <0.3.0-0",
+    "@deepseek-ai/dsh-tools": ">=0.1.5-rc.2 <0.2.0 || >=0.2.0-rc.1 <0.3.0-0 || >=0.2.1-0 <0.3.0-0"
   },
   "dependencies": {
     "@deepseek-ai/schemastery": "^3.18.2"