chnages

app/agents/context_agent.py

@@ -16,7 +16,7 @@ context_agent = create_agent(
     middleware=[
         ToolCallLimitMiddleware[Any,None](
            tool_name="search_memory",
-           run_limit=7,
+           run_limit=5,
          thread_limit=10,
         )
    ] ,

app/core/config.py

@@ -5,24 +5,16 @@ BASE_DIR = Path(__file__).resolve().parent.parent.parent
  
 class Settings(BaseSettings):
     PROJECT_NAME: str = "Email Assistant Project"
- 
     GROQ_API_KEY: str
-    # PINECONE_API_KEY: str
-
-    # CLOUDINARY_CLOUD_NAME: str
-    # CLOUDINARY_API_KEY: str
-    # CLOUDINARY_API_SECRET: str
-
-    
-    # DB_URL_FOR_CHECKPOINTER_STORE: str
+    DB_URL_FOR_CHECKPOINTER_STORE: str
 
-    # DB_URL_FOR_SQL_AL:str
+    DB_URL_FOR_SQL_AL:str
  
-    # model_config = SettingsConfigDict(
-    #     env_file=str(BASE_DIR / ".env"),
-    #     env_file_encoding="utf-8",
-    #     extra="ignore"
-    # )
+    model_config = SettingsConfigDict(
+        env_file=str(BASE_DIR / ".env"),
+        env_file_encoding="utf-8",
+        extra="ignore"
+    )
  
 settings = Settings()
 

app/database/connection.py

@@ -1,6 +1,9 @@
 from sqlalchemy import create_engine
 from sqlalchemy.orm import sessionmaker, Session
-from app.core.config import DB_URL_FOR_SQL_AL
+from app.core.config import settings
+
+DB_URL_FOR_SQL_AL=settings.DB_URL_FOR_SQL_AL
+
 engine = create_engine(
     DB_URL_FOR_SQL_AL, 
     echo=False, 

app/nodes/check_token_count_node.py

@@ -1,14 +1,7 @@
 from app.state.state import EmailAgentState
 from langchain_groq import ChatGroq
+from app.utils.token_utils import count_input_tokens
 
-llm_for_token_count=ChatGroq(
-    model="meta-llama/llama-4-scout-17b-16e-instruct",
-    temperature=0.1,
-)
-
-def count_input_tokens(subject:str ,body:str) -> int:
-    text=subject+body
-    return llm_for_token_count.get_num_tokens(text)
 
 def check_token_count_node(state: EmailAgentState):
     """

app/nodes/store_memory_data_node.py

@@ -6,6 +6,8 @@ from app.state.state import EmailAgentState
 from app.agents.memory_manager_agent import memory_manager_agent 
 from app.database.connection import get_session
 from app.database.utils import save_sent_email, save_received_email
+from langchain_core.messages import AIMessage
+from app.utils.token_utils import *
 
 def store_memory_and_data_node(state: EmailAgentState, config: RunnableConfig):
     """
@@ -13,21 +15,38 @@ def store_memory_and_data_node(state: EmailAgentState, config: RunnableConfig):
     """
     print("--- Memory Node: Persisting interaction to DB ---")
 
+
+    body_for_memory_agent=state.get("reply_email_body")
+    reply_subject=state.get("reply_subject")
+    
+
+    if count_input_tokens(body_for_memory_agent,reply_subject)>100000:
+        body_for_memory_agent=summarise_email_body(body_for_memory_agent)
+
     # 1. Prepare the memory summary
     prompt = memory_agent_template.invoke({
         "user_name": state["user_name"],
         "senders_email_id": state["sender_email_id"],
         "user_email_id": state["user_email_id"],
-        "sent_email_body": state.get("reply_email_body", "No response found"), # Use .get() for safety
+        "sent_email_body": body_for_memory_agent,
         "incoming_email_body": state["sender_email_body"],
     })
 
     # 2. Invoke memory agent logic
-    memory_manager_agent.invoke(
+     
+
+    print("invoking memory manager")
+
+    
+    response = memory_manager_agent.invoke(
         {"messages": prompt.to_messages()},
         config=config
     )
 
+    print(response)
+
+    memory_stored_summary = response[0]['value']['content']['summary']
+
     # 3. Robust Database Operations
     sender_id = state['user_id']
     thread_id = config.get("configurable", {}).get("thread_id")
@@ -44,4 +63,4 @@ def store_memory_and_data_node(state: EmailAgentState, config: RunnableConfig):
             print(f"--- Memory Node Error: {e} ---")
             raise e
 
-    return {"memory_agent_messages": [AIMessage(content="Memory successfully persisted.")],"email_sent": True}
+    return {"memory_agent_messages": [AIMessage(content=memory_stored_summary)],"email_sent": True}

app/nodes/summarise_email_body_node.py

@@ -5,22 +5,9 @@ from langchain_classic.text_splitter import RecursiveCharacterTextSplitter
 from langchain_core.documents import Document
 from langchain_classic.prompts import PromptTemplate
 from app.agents.summarizer_agent import summarizer_agent
-from app.prompts.summarizer_agent_prompt import summarize_agent_initial_prompt_template,summarize_agent_refine_template
-def summarise_email_body(body: str):
-    # Tip: 100 is very small for an email; 500-1000 is usually better
-    text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
-    docs = [Document(page_content=t) for t in text_splitter.split_text(body)]
-
-    chain = load_summarize_chain(
-        llm=summarizer_agent, 
-        chain_type="refine",
-        question_prompt=summarize_agent_initial_prompt,
-        refine_prompt=summarize_agent_refine_prompt,
-        document_variable_name="body"  # <--- ADD THIS LINE
-    )
-    summary = chain.invoke(docs)
-    return summary['output_text']
-
+from langchain_groq import ChatGroq
+from app.prompts.summarizer_agent_prompt import *
+from app.utils.token_utils import summarise_email_body
 
 def summarise_email_body_node(state:EmailAgentState)->dict:
 

app/prompts/email_writing_agent_prompt.py

@@ -24,11 +24,14 @@ Output: "Stored."
 <rules>
 1. NEW DRAFT: Call `create_gmail_draft` first.
 2. REJECTION: If "DRAFT REJECTED", rewrite and call `create_gmail_draft_tool` again.
-3. SENDING: Only call `send_draft_by_id` when user explicitly orders to send. **dont call send_draft_by_id until user  allows you to send the draft**
+3. SENDING: Only call `send_draft_by_id` when user explicitly orders to send. 
 4. ARCHIVING: After `send_draft_by_id` returns a Message ID.
 important:You are not allowed to send until the user explicitly orders to send.
 5.Use **{sender_email_id}** as the recipient not the name.
+**dont call send_draft_by_id until user  allows you to send the draft**
 </rules>
+     
+
 
 """)
 ])

app/state/state.py

@@ -61,7 +61,7 @@ class EmailAgentState(TypedDict):
 
     email_sent: Optional[bool]
 
-   
+    
     human_approved: Optional[bool]
     reply_email_body:Optional[str]  
 

app/utils/token_utils.py

@@ -0,0 +1,30 @@
+from langchain_groq import ChatGroq
+from langchain_classic.chains.summarize import load_summarize_chain
+from langchain_classic.text_splitter import RecursiveCharacterTextSplitter
+from langchain_core.documents import Document
+from app.agents.summarizer_agent import summarizer_agent
+from app.prompts.summarizer_agent_prompt import *
+
+llm_for_token_count=ChatGroq(
+    model="meta-llama/llama-4-scout-17b-16e-instruct",
+    temperature=0.1,
+)
+
+def count_input_tokens(subject:str ,body:str) -> int:
+    text=subject+body
+    return llm_for_token_count.get_num_tokens(text)
+
+def summarise_email_body(body: str):
+    # Tip: 100 is very small for an email; 500-1000 is usually better
+    text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
+    docs = [Document(page_content=t) for t in text_splitter.split_text(body)]
+
+    chain = load_summarize_chain(
+        llm=summarizer_agent, 
+        chain_type="refine",
+        question_prompt=summarize_agent_initial_prompt,
+        refine_prompt=summarize_agent_refine_prompt,
+        document_variable_name="body"  # <--- ADD THIS LINE
+    )
+    summary = chain.invoke(docs)
+    return summary['output_text']