diff --git a/lib/air_runtime/modes/arc_mode.py b/lib/air_runtime/modes/arc_mode.py
index 0651334..c926249 100644
--- a/lib/air_runtime/modes/arc_mode.py
+++ b/lib/air_runtime/modes/arc_mode.py
@@ -249,8 +249,11 @@ def parallel_review_mode(project_root: Path, todo_path: Path) -> dict:
         markdown_lines.append(f"- `{g.name}`: {', '.join(g.task_ids)} — {g.reason}")
     paths["execution_plan_md"].write_text("\n".join(markdown_lines) + "\n", encoding="utf-8")
 
-    atomic_json_write(paths["state"], {"enabled": True, "updatedAt": now_iso(),
-                                        "projectRoot": str(project_root)})
+    # 保留 arcPhase 字段，避免 phase 被 reset 为 discussing
+    existing_state = safe_json_load(paths["state"]) or {}
+    state_payload = {**existing_state, "enabled": True, "updatedAt": now_iso(),
+                     "projectRoot": str(project_root)}
+    atomic_json_write(paths["state"], state_payload)
     return {"json_path": str(review_json_path), "markdown_path": str(review_md_path),
             "execution_plan_json_path": str(paths["execution_plan_json"]),
             "parallel_group_count": len(review.parallel_groups),
diff --git a/lib/air_runtime/modes/do_mode.py b/lib/air_runtime/modes/do_mode.py
index 9bc7d7d..7aed2a8 100644
--- a/lib/air_runtime/modes/do_mode.py
+++ b/lib/air_runtime/modes/do_mode.py
@@ -244,7 +244,8 @@ def main(args) -> None:
         print(f"active_task_id={s['activeTaskId']}")
         print(f"known_tasks={','.join(s['taskIds'])}")
     elif sub == "enter":
-        result = enter_worker(project_root, tid)
+        task_text = getattr(args, "task_text", "") or ""
+        result = enter_worker(project_root, tid, task_text=task_text)
         print("airplan_mode=do")
         print(f"task_id={result['taskId']}")
         print(f"brief_path={result['briefPath']}")
diff --git a/lib/air_runtime/todo_parser.py b/lib/air_runtime/todo_parser.py
index c04c1ec..f70b4b5 100644
--- a/lib/air_runtime/todo_parser.py
+++ b/lib/air_runtime/todo_parser.py
@@ -69,9 +69,13 @@ def parse_tasks(todo_path: Path) -> list[TodoTask]:
             continue
 
         task_cell = cells[col_map.get("task", 0)] if col_map.get("task", 0) < len(cells) else ""
-        # 提取 [T-xxx] 格式的任务 ID
+        # 优先提取 [T-xxx] 方括号格式的 ID；若没有则尝试从开头提取 G-001/H-000 类短 ID
         tid_match = re.match(r"\[([A-Za-z0-9_\-\.]+)\]", task_cell)
-        task_id = tid_match.group(1) if tid_match else task_cell
+        if tid_match:
+            task_id = tid_match.group(1)
+        else:
+            short_match = re.match(r"^([A-Z]+-\d+[a-z]*)", task_cell)
+            task_id = short_match.group(1) if short_match else task_cell
         task = task_cell
         status = cells[col_map.get("status", 1)] if col_map.get("status", 1) < len(cells) else "TODO"
         files_dirs = cells[col_map.get("files_dirs", 2)] if col_map.get("files_dirs", 2) < len(cells) else ""
diff --git a/scripts/airplan.py b/scripts/airplan.py
index 91c0447..c09b8f5 100644
--- a/scripts/airplan.py
+++ b/scripts/airplan.py
@@ -170,6 +170,8 @@ def parse_args(argv: list[str] | None = None) -> argparse.Namespace:
                        help="截图后端偏好（xdb 使用）: kms|xvfb|fallback|auto")
     parser.add_argument("--output", default="",
                        help="截图输出文件名（xdb 使用，默认 screenshot.png）")
+    parser.add_argument("--task-text", default="",
+                       help="任务描述文本（do 使用，用于 UI 任务检测和路由）")
     return parser.parse_args(argv)