feat: make other agents support asking user input in MessageAction. (All-Hands-AI#1777)

* feat: make other agents support asking user input in MessageAction.

* Update agenthub/micro/_instructions/actions/message.md

Co-authored-by: Robert Brennan <[email protected]>

* Update agenthub/micro/_instructions/actions/message.md

Co-authored-by: Robert Brennan <[email protected]>

* feat: make other agents support asking user input in MessageAction.

* Regenerate test artifacts

---------

Co-authored-by: aaren.xzh <[email protected]>
Co-authored-by: Robert Brennan <[email protected]>
Co-authored-by: Boxuan Li <[email protected]>

Author: 4 people

agenthub/micro/_instructions/actions/message.md

@@ -1,2 +1,3 @@
-* `message` - make a plan, set a goal, or record your thoughts. Arguments:
+* `message` - make a plan, set a goal, record your thoughts, or ask for more input from the user. Arguments:
   * `content` - the thought to record
+  * `wait_for_response` - set to `true` to wait for the user to respond before proceeding

agenthub/monologue_agent/utils/prompts.py

@@ -44,8 +44,9 @@
   * `branch` - the name of the branch to push
 * `recall` - recalls a past memory. Arguments:
   * `query` - the query to search for
-* `message` - make a plan, set a goal, or record your thoughts. Arguments:
+* `message` - make a plan, set a goal, record your thoughts, or ask for more input from the user. Arguments:
   * `content` - the message to record
+  * `wait_for_response` - set to `true` to wait for the user to respond before proceeding
 * `finish` - if you're absolutely certain that you've completed your task and have tested your work, use the finish action to stop working.
 
 %(background_commands)s

agenthub/planner_agent/prompt.py

@@ -82,8 +82,9 @@
   * `command_id` - the ID of the background command to kill
 * `browse` - opens a web page. Arguments:
   * `url` - the URL to open
-* `message` - make a plan, set a goal, or record your thoughts. Arguments:
+* `message` - make a plan, set a goal, record your thoughts, or ask for more input from the user. Arguments:
   * `content` - the message to record
+  * `wait_for_response` - set to `true` to wait for the user to respond before proceeding
 * `add_task` - add a task to your plan. Arguments:
   * `parent` - the ID of the parent task (leave empty if it should go at the top level)
   * `goal` - the goal of the task

tests/integration/mock/MonologueAgent/test_edits/prompt_001.log

@@ -360,8 +360,9 @@ Here are the possible actions:
   * `branch` - the name of the branch to push
 * `recall` - recalls a past memory. Arguments:
   * `query` - the query to search for
-* `message` - make a plan, set a goal, or record your thoughts. Arguments:
+* `message` - make a plan, set a goal, record your thoughts, or ask for more input from the user. Arguments:
   * `content` - the message to record
+  * `wait_for_response` - set to `true` to wait for the user to respond before proceeding
 * `finish` - if you're absolutely certain that you've completed your task and have tested your work, use the finish action to stop working.
 
 

tests/integration/mock/MonologueAgent/test_edits/prompt_002.log

@@ -334,20 +334,21 @@ This is your internal monologue, in JSON format:
   },
   {
     "source": "agent",
-    "action": "read",
+    "action": "run",
     "args": {
-      "path": "bad.txt",
-      "start": 0,
-      "end": -1,
+      "command": "pwd",
+      "background": false,
       "thought": ""
     }
   },
   {
     "source": "agent",
-    "observation": "read",
-    "content": "This is a stupid typoo.\nReally?\nNo mor typos!\nEnjoy!\n",
+    "observation": "run",
+    "content": "/workspace",
     "extras": {
-      "path": "bad.txt"
+      "command_id": -1,
+      "command": "pwd",
+      "exit_code": 0
     }
   }
 ]
@@ -378,8 +379,9 @@ Here are the possible actions:
   * `branch` - the name of the branch to push
 * `recall` - recalls a past memory. Arguments:
   * `query` - the query to search for
-* `message` - make a plan, set a goal, or record your thoughts. Arguments:
+* `message` - make a plan, set a goal, record your thoughts, or ask for more input from the user. Arguments:
   * `content` - the message to record
+  * `wait_for_response` - set to `true` to wait for the user to respond before proceeding
 * `finish` - if you're absolutely certain that you've completed your task and have tested your work, use the finish action to stop working.
 
 

tests/integration/mock/MonologueAgent/test_edits/prompt_003.log

@@ -334,39 +334,40 @@ This is your internal monologue, in JSON format:
   },
   {
     "source": "agent",
-    "action": "read",
+    "action": "run",
     "args": {
-      "path": "bad.txt",
-      "start": 0,
-      "end": -1,
+      "command": "pwd",
+      "background": false,
       "thought": ""
     }
   },
   {
     "source": "agent",
-    "observation": "read",
-    "content": "This is a stupid typoo.\nReally?\nNo mor typos!\nEnjoy!\n",
+    "observation": "run",
+    "content": "/workspace",
     "extras": {
-      "path": "bad.txt"
+      "command_id": -1,
+      "command": "pwd",
+      "exit_code": 0
     }
   },
   {
     "source": "agent",
-    "action": "write",
+    "action": "run",
     "args": {
-      "path": "bad.txt",
-      "content": "This is a stupid typo.\nReally?\nNo more typos!\nEnjoy!\n",
-      "start": 0,
-      "end": -1,
+      "command": "ls",
+      "background": false,
       "thought": ""
     }
   },
   {
     "source": "agent",
-    "observation": "write",
-    "content": "",
+    "observation": "run",
+    "content": "bad.txt",
     "extras": {
-      "path": "bad.txt"
+      "command_id": -1,
+      "command": "ls",
+      "exit_code": 0
     }
   }
 ]
@@ -397,8 +398,9 @@ Here are the possible actions:
   * `branch` - the name of the branch to push
 * `recall` - recalls a past memory. Arguments:
   * `query` - the query to search for
-* `message` - make a plan, set a goal, or record your thoughts. Arguments:
+* `message` - make a plan, set a goal, record your thoughts, or ask for more input from the user. Arguments:
   * `content` - the message to record
+  * `wait_for_response` - set to `true` to wait for the user to respond before proceeding
 * `finish` - if you're absolutely certain that you've completed your task and have tested your work, use the finish action to stop working.