Updated the tools

my_agent/utils/tools.py

@@ -7,6 +7,10 @@ import numpy as np
 from langchain_core.tools import tool
 from .data_loader import load_influencer_data
 from .models_loader import  ST , llm
+from sklearn.metrics.pairwise import cosine_similarity
+import numpy as np
+from langchain_core.messages import SystemMessage
+import re
 
 
 os.environ['GROQ_API_KEY']=os.getenv('GROQ_API_KEY')
@@ -30,26 +34,23 @@ class BrainstromTopicFormatter(BaseModel):
   topic4:str=Field(description="Fourth brainstorming topic of the story")
 
 class QueryFormatter(BaseModel):
-    idea:str = Field(description="Any idea or query about the business.")
+    messages:str = Field(description="The user query")
     business_details: str = Field(description="The details of the business of that user.")
 
 @tool("influencer's data-retrieval-tool", args_schema=QueryFormatter, return_direct=False,description="Retrieve influencer-related data for a given query.")
-def retrieve_tool(idea, business_details):
-    
-    # """This tool is responsible for the retrieval of the influencer's data using semantic search by reading any **idea or query about the business** and the **business details of the user.**
-    #  But remember, the idea have to be valid first. Don't retrieve anything if the idea is invalid or it is like General Question Answering or follow up questions.
-    #  If you find the idea as invalid, write the value as "None" in the idea so that i can process it."""
-    
-    """This tool is responsible for the retrieval of the influencer's data using semantic search by reading any **idea or query about the business** and the **business details of the user.**
-     ."""
-    
-
-    embedded_query = ST.encode(str(idea)+str(business_details))  # Embed each topic
+def retrieve_tool(messages, business_details):
+    '''Always invoke this tool once.'''
+    print('The query for retrieval is:',messages)
+    embedded_query = ST.encode(str(messages)+str(business_details))  # Embed each topic
     data = load_influencer_data()
-    scores, retrieved_examples = data.get_nearest_examples("embeddings", embedded_query, k=3)
+    scores, retrieved_examples = data.get_nearest_examples("embeddings", embedded_query, k=2)
 
     # Construct a list of dictionaries for this topic
     result = [{user: story} for user, story in zip(retrieved_examples['username'], retrieved_examples['agentic_story'])]
-    # result = [{u: {"story": s, "likes": l, "comments": c}} for u, s, l, c in zip(retrieved_examples['username'], retrieved_examples['agentic_story'], retrieved_examples['likes'], retrieved_examples['comments'])]
-    print('The tool response:',result)
+    print('Tool response:',result)
+
+
     return result
+
+
+