fix(ack): keep coordinator loop active after dispatch

tests/test_ack_skill.py

@@ -121,6 +121,30 @@ class AckSkillContentTests(unittest.TestCase):
         self.assertIn("validation_ready", delivery)
         self.assertIn("访问地址和用户下一步", delivery)
 
+    def test_coordinator_keeps_looping_and_chains_developer_to_test(self) -> None:
+        skill = (REPO_ROOT / "skills" / "ack" / "SKILL.md").read_text(
+            encoding="utf-8"
+        )
+
+        required = (
+            "这不是 fire-and-forget",
+            "不得在 `launch` / `dispatch` 后只回复“已派发”并结束当前会话",
+            "`check --wait`",
+            "`worker_probe.py`",
+            "Developer 回报 `worker_done` 或 `escalation`",
+            "写回 `fixed_by_dev`",
+            "主动派发 Test",
+            "继续等待 `retest_result`",
+        )
+        for expected in required:
+            self.assertIn(expected, skill)
+
+        dispatch_pos = skill.index("主动派发 Test")
+        dev_done_pos = skill.index("Developer 回报 `worker_done` 或 `escalation`")
+        test_done_pos = skill.index("继续等待 `retest_result`")
+        self.assertLess(dev_done_pos, dispatch_pos)
+        self.assertLess(dispatch_pos, test_done_pos)
+
     def test_ack_knowledge_resources_and_version_are_present(self) -> None:
         ack_dir = REPO_ROOT / "skills" / "ack"