updated the context calculation

app/api/endpoints/openai.py

@@ -49,9 +49,14 @@ async def openai_chat(
             }
 
         # Perform Async call to OpenAI
+        messages = []
+        if chat_data.context:
+            messages.append({"role": "system", "content": chat_data.context})
+        messages.append({"role": "user", "content": chat_data.prompt})
+
         response = await client.chat.completions.create(
             model="gpt-3.5-turbo",
-            messages=[{"role": "user", "content": chat_data.prompt}]
+            messages=messages
         )
         
         # Track usage
@@ -61,7 +66,16 @@ async def openai_chat(
                 module.ingress_tokens += usage.prompt_tokens
                 module.egress_tokens += usage.completion_tokens
                 module.total_tokens += usage.total_tokens
-                db.commit()
+            else:
+                # Fallback estimation
+                total_content = "".join([m["content"] for m in messages])
+                prompt_tokens = len(total_content) // 4
+                completion_tokens = len(response.choices[0].message.content) // 4
+                module.ingress_tokens += prompt_tokens
+                module.egress_tokens += completion_tokens
+                module.total_tokens += (prompt_tokens + completion_tokens)
+            
+            db.commit()
 
         return {
             "status": "success",