feat: support key combinations in press_key function and update related tests

SKILL.md

@@ -10,6 +10,7 @@ ScreenJob lets an agent execute tasks that require a real desktop UI plus termin
 - Mouse/keyboard control (`click`, `type`, `press_key`)
 - Terminal execution (`execute_command`, `sleep`)
 - Structured completion payload (`task_complete(return=..., data=...)`)
+- Automatic final verification screen capture on completion
 - Safety gate, auth, history, and live monitoring
 
 ## Important Environment Note
@@ -30,7 +31,12 @@ Agents can use ScreenJob to launch and control GUI workflows, including orchestr
 
 1. Submit job via CLI or API.
 2. Agent performs tool loop.
-3. Read final `response.return` and `response.data` from job status.
+3. Read final `response.return`, `response.data`, and `verification` from job status.
+
+Keyboard combo rule:
+
+- For shortcuts, use one `press_key` call with combo syntax, for example: `win+r`, `ctrl+shift+esc`.
+- Do not split modifier combos into separate calls.
 
 ## API Quick Reference
 
@@ -79,10 +85,18 @@ Result contract in job payload:
   "status": "completed",
   "response": {
     "return": "Task completed successfully",
-    "data": "file1.txt\nfile2.txt"
+    "data": "file1.txt\nfile2.txt",
+    "verification": {
+      "ok": true,
+      "path": "C:/.../screens/screen_final_verification_step_006.png"
+    }
   },
   "return": "Task completed successfully",
-  "data": "file1.txt\nfile2.txt"
+  "data": "file1.txt\nfile2.txt",
+  "verification": {
+    "ok": true,
+    "path": "C:/.../screens/screen_final_verification_step_006.png"
+  }
 }
 ```