collect data tool

src/sayvai_tools/tools/__init__.py

@@ -28,6 +28,7 @@
 from sayvai_tools.tools.youtube.comments import (ListCommentRepliesTool,
                                                  ReplyToCommentTool)
 from sayvai_tools.tools.youtube.utils import get_youtube_credentials
+from sayvai_tools.tools.collect_data_from_user import CollectUserDataTool, create_data_model
 
 __all__: List[str] = [
     "ConversationalHuman",
@@ -54,6 +55,8 @@
     "ListCommentRepliesTool",
     "ReplyToCommentTool",
     "load_tools",
+    "CollectUserDataTool",
+    "create_data_model"
 ]
 
 

src/sayvai_tools/tools/collect_data_from_user/__init__.py

@@ -0,0 +1,6 @@
+from sayvai_tools.tools.collect_data_from_user.collect_data_from_user import CollectUserDataTool, create_data_model
+
+__all__ = [
+    "CollectUserDataTool",
+    "create_data_model"
+]

src/sayvai_tools/tools/collect_data_from_user/collect_data_from_user.py

@@ -0,0 +1,32 @@
+from typing import Type, Optional, Dict
+
+from langchain_core.callbacks import CallbackManagerForToolRun
+from langchain_core.tools import BaseTool
+from pydantic import BaseModel, create_model, Field
+
+
+def create_data_model(fields: Dict[str, str]) -> Type[BaseModel]:
+    """
+    Dynamically creates a Pydantic model from the fields dictionary.
+
+    :param fields: A dictionary where keys are field names and values are descriptions.
+    :return: A dynamically created Pydantic model.
+    """
+    field_definitions = {field: (str, Field(description=desc)) for field, desc in fields.items()}
+    return create_model('UserData', **field_definitions)
+
+
+class CollectUserDataTool(BaseTool):
+    name: str = "user_data_collector"
+    description: str = "Collects data from the user based on specified fields."
+
+    args_schema: Type[BaseModel]
+
+    def _run(self, run_manager: Optional[CallbackManagerForToolRun] = None, **kwargs) -> bool:
+        try:
+            # Validate data using the dynamically created model
+            print("Collected and validated data:", kwargs)
+            return True
+        except Exception as e:
+            print("Error in data collection or validation:", str(e))
+            return False