From 81f654c3fdeb0ac7b706ab8330d276b0c50f7458 Mon Sep 17 00:00:00 2001
From: Paulo Reyes <aijin02@gmail.com>
Date: Tue, 10 Feb 2026 03:28:46 +0800
Subject: [PATCH] new endpoint settings

---
 app/api/endpoints/gemini.py | 24 ++++++++++++++++++++++--
 app/core/prompts.py         |  5 +++++
 2 files changed, 27 insertions(+), 2 deletions(-)
 create mode 100644 app/core/prompts.py

diff --git a/app/api/endpoints/gemini.py b/app/api/endpoints/gemini.py
index da9cb38..e0dfccb 100644
--- a/app/api/endpoints/gemini.py
+++ b/app/api/endpoints/gemini.py
@@ -8,12 +8,20 @@ from app.core.config import settings
 from pydantic import BaseModel
 from google import genai
 import asyncio
+from google.genai import types
+from app.core.prompts import GEMINI_SYSTEM_PROMPT
 
 router = APIRouter()
 
 class LLMRequest(BaseModel):
     prompt: str
     context: str = ""
+    system_prompt: str | None = None
+    knowledge_base: str | None = None
+    temperature: float = 0.7
+    top_p: float = 0.95
+    top_k: int = 40
+    max_output_tokens: int = 8192
 
 # Shared client instance (global)
 _client = None
@@ -58,11 +66,23 @@ async def gemini_chat(
         if chat_data.context:
             prompt_content = f"Context: {chat_data.context}\n\nPrompt: {chat_data.prompt}"
 
+        # Prepare system instruction
+        system_instruction = chat_data.system_prompt or GEMINI_SYSTEM_PROMPT
+        if chat_data.knowledge_base:
+            system_instruction += f"\n\nKnowledge Base:\n{chat_data.knowledge_base}"
+
         # Using the async generation method provided by the new google-genai library
         # We use await to ensure we don't block the event loop
         response = await client.aio.models.generate_content(
-            model="gemini-2.0-flash", 
-            contents=prompt_content
+            model="gemini-2.5-flash", 
+            contents=prompt_content,
+            config=types.GenerateContentConfig(
+                system_instruction=system_instruction,
+                temperature=chat_data.temperature,
+                top_p=chat_data.top_p,
+                top_k=chat_data.top_k,
+                max_output_tokens=chat_data.max_output_tokens
+            )
         )
         
         # Track usage if valid module
diff --git a/app/core/prompts.py b/app/core/prompts.py
new file mode 100644
index 0000000..c1dcf9b
--- /dev/null
+++ b/app/core/prompts.py
@@ -0,0 +1,5 @@
+
+# System Prompts Configuration
+
+GEMINI_SYSTEM_PROMPT = """You are LDEx AI, a helpful AI assistant. 
+Answer the user's questions concisely and accurately based on the knowledge base provided."""