fix date picker issue (#1523)

skyvern/forge/agent.py

@@ -846,6 +846,19 @@ class ForgeAgent:
                         action=action,
                         action_result=results,
                     )
+                    if results[-1].skip_remaining_actions:
+                        LOG.warning(
+                            "Going to stop executing the remaining actions",
+                            task_id=task.task_id,
+                            step_id=step.step_id,
+                            step_order=step.order,
+                            step_retry=step.retry_index,
+                            action_idx=action_idx,
+                            action=action,
+                            action_result=results,
+                        )
+                        break
+
                 elif results and isinstance(action, DecisiveAction):
                     LOG.warning(
                         "DecisiveAction failed, but not stopping execution and not retrying the step",

skyvern/forge/prompts/skyvern/custom-select.j2

@@ -1,19 +1,21 @@
-You are performing a {{ "multi-level selection" if select_history else "selection" }} action on an HTML page. Assist the user in selecting the most appropriate option(or typing some values to search if neccesary) to advance toward their goal, considering the context, user details, and the DOM elements provided in the list.
+You are performing a {{ "multi-level selection" if select_history else "selection" }} action on an HTML page. Assist the user in selecting the most appropriate option(or typing some values to search if neccesary) to advance toward their goal, considering the context, user details, and the HTML elements provided in the list.
 
 You can identify the matching element based on the following guidelines:
   1. Select the most suitable element based on the user goal, user details, and the context.
-  2. If none of the options perfectly match, and there is no search box for input, but there is a fallback option such as "Others" or "None of the above" in the DOM elements, you can consider it a match.
+  2. If none of the options perfectly match, and there is no search box for input, but there is a fallback option such as "Others" or "None of the above" in the HTML elements, you can consider it a match.
   3. If a field is required, do not leave it blank.
   4. If a field is required, do not select a placeholder value, such as "Please select", "-", or "Select...".
   5. Exclude loading indicators like "loading more results" as valid options.{% if select_history %}
-  6. The selection history displays the previously selected values for the multi-level selection. Continue to complete the entire selection process.{% endif %}
+  6. The selection history displays the previously selected values for the multi-level selection. Continue to complete the entire selection process.{% if is_date_related %}
+  7. Date picker might be triggered, you goal is to set the correct start date and end date.{% endif %}{% endif %}
 
 MAKE SURE YOU OUTPUT VALID JSON. No text before or after JSON, no trailing commas, no comments (//), no unnecessary quotes, etc.
 Each interactable element is tagged with an ID.
 
 Reply in JSON format with the following keys:
 {
-    "reasoning": str, // The reasoning behind the action. Be specific, referencing the value and the element id in your reasoning. Mention why you chose the element id. Keep the reasoning short and to the point.
+    "page_info": str, // Think step by step. Describe the page information you parsed from the HTML elements. Your action should be based on the current page information.
+    "reasoning": str, // The reasoning behind the current single action. Be specific, referencing the value and the element id in your reasoning. Mention why you chose the element id. Keep the reasoning short and to the point.
     "confidence_float": float, // The confidence of the action. Pick a number between 0.0 and 1.0. 0.0 means no confidence, 1.0 means full confidence
     "id": str, // The id of the element to take action on. The id has to be one from the elements list
     "action_type": str, // It's a string enum: "CLICK", "INPUT_TEXT". "CLICK" is an option you'd like to click to choose. "INPUT_TEXT" is an element you'd like to input text into for searching, but it only should be used when there's no valid option to click.

skyvern/forge/prompts/skyvern/opened-dropdown-confirm.j2

@@ -3,6 +3,7 @@ There is a screenshot from a part of a web HTML page. Help me confirm if it is a
 An open dropdown menu can be defined as:
   - At least one option is visible in the screenshot.
   - A calendar view could be considered as an open dropdown menu. But DO NOT consider an calendar icon as the dropdown menu.
+  - If the screenshot contains multiple input fields for selecting quantities, consider as the dropdown menu.
   - Do not consider it an open dropdown menu if the only visible option displays a message like "No results" or "No match".
   - Do not consider it an open dropdown menu if the only visible element displays a placeholder like "Please select", "-", or "Select...".
 

skyvern/forge/prompts/skyvern/parse-input-or-select-context.j2

@@ -9,6 +9,7 @@ Reply in the following JSON format:
     "is_required": bool, // True if this is a required field, otherwise false.
     "is_search_bar": bool, // True if the element to take the action is a search bar, otherwise false.
     "is_location_input": bool, // True if the element is asking user to input where he lives, otherwise false. For example, it is asking for location, or address, or other similar information. Output False if it only requires ZIP code or postal code.
+    "is_date_related": bool, // True if the field is related to date input or select, otherwise false.
 }
 
 Existing reasoning context: