refactor(core): task_output 工具使用外部描述文件

packages/core/src/tools/descriptions/task/task_output.txt

@@ -0,0 +1,7 @@
+- Retrieves output from a running or completed task (background shell, agent, or remote session)
+- Takes a task_id parameter identifying the task
+- Returns the task output along with status information
+- Use block=true (default) to wait for task completion
+- Use block=false for non-blocking check of current status
+- Task IDs can be found using the /tasks command
+- Works with all task types: background shells, async agents, and remote sessions

packages/core/src/tools/task/task_output.ts

@@ -1,5 +1,6 @@
 import type { ToolWithMetadata } from '../types.js';
 import { getAgentManager } from '../../agent/manager.js';
+import { loadDescription } from '../load_description.js';
 
 /**
  * TaskOutput 工具
@@ -7,15 +8,7 @@ import { getAgentManager } from '../../agent/manager.js';
  */
 export const taskOutputTool: ToolWithMetadata = {
   name: 'task_output',
-  description: `获取后台运行的 Agent 执行结果。
-
-当使用 task 工具的 run_in_background 参数启动后台 Agent 后，
-使用此工具查询执行状态和结果。
-
-使用示例:
-- 查询结果（阻塞等待）: task_output({ task_id: "abc123" })
-- 检查状态（不阻塞）: task_output({ task_id: "abc123", block: false })
-- 设置超时: task_output({ task_id: "abc123", timeout: 60000 })`,
+  description: loadDescription('task_output'),
   parameters: {
     task_id: {
       type: 'string',