module references clean

modules/workflow/managerChat.py

@@ -159,6 +159,8 @@ class ChatManager:
                 "duration": (datetime.now(UTC) - datetime.fromisoformat(workflow.startedAt)).total_seconds()
             }
             
+            # Get user language from service
+            userLanguage = self.service.user.language
             
             # Prepare messages for AI context
             messages = [
@@ -188,7 +190,7 @@ class ChatManager:
                     })
             
             # Generate feedback using AI
-            feedback = await self.callAi(messages, produceUserAnswer=True, temperature=0.7)
+            feedback = await self.service.callAi(messages, produceUserAnswer=True, temperature=0.7)
             
             return feedback
             

notes/changelog.txt

@@ -1,23 +1,4 @@
-- all AI calls to route over self service (AI basic, ai special, ai...) and to streamline
+Prompt for task definition:
-
-- integrate, then remove:
-    callAi(prompt, context) -> str
-    callAi4Image(prompt, context, imageData)-> str
-    _callAiBasic
-    _callAiAdvanced
-
-in manageChat
-- not to manipulate workflow or messages --> to use functions from servicechatclass
-- to add CRUD for tasks to workflow
-- to move methods to methods class
-
-- service object to define correctly
-- folder workflow modules --> solve problems types
-
-
-Procedure for task generation:
-
-- prompt input:
   - original user input (summary)
   - prompt for task to do
   - method list
@@ -33,12 +14,10 @@ Procedure for task generation:
       - as parameter only to use available items of documentReference and connectionReference or resultLabel of a previous method.action
     - required result format: json for TaskResult, nothing else
 
-- result parsing:
-  - execute task actions by using data references stepwise, all documents in TaskResults to save as ChatDocuments, to be available for next action.
 
-
+TODO
-TODO:
+- result parsing: execute task actions by using data references stepwise, all documents in TaskResults to save as ChatDocuments, to be available for next action.
-- action output filename to be available as TaskResult with name to reference object and ChatDocument list
+- action execution: to use conversion functions, user DocumentReference and ConnectionReference
 
 
 service (ServiceContainer)