From dabce8b6fb40ab441951ecdc8b049e9b19476d08 Mon Sep 17 00:00:00 2001
From: kswdev0 <skawksmsduzo@gmail.com>
Date: Fri, 27 Mar 2026 10:59:43 +0900
Subject: [PATCH] refactor: replace Ollama client with HTTPX for API requests
 in AIPlanner

---
 ai_planner.py | 26 +++++++++++++++++---------
 1 file changed, 17 insertions(+), 9 deletions(-)

diff --git a/ai_planner.py b/ai_planner.py
index c7bafa0..fb1d655 100644
--- a/ai_planner.py
+++ b/ai_planner.py
@@ -19,7 +19,7 @@ import sys
 import time
 import threading
 import traceback
-import ollama
+import httpx
 
 
 OLLAMA_MODEL = os.environ.get("OLLAMA_MODEL", "qwen3.5:9b")
@@ -192,14 +192,13 @@ class AIPlanner:
         spinner.start()
 
         try:
-            client = ollama.Client(host=OLLAMA_HOST)
-            response = client.chat(
-                model=OLLAMA_MODEL,
-                messages=[
+            payload = {
+                "model": OLLAMA_MODEL,
+                "messages": [
                     {"role": "system", "content": SYSTEM_PROMPT},
-                    {"role": "user",   "content": "/no_think\n" + user_message},
+                    {"role": "user",   "content": user_message},
                 ],
-                format={
+                "format": {
                     "type": "object",
                     "properties": {
                         "thinking":     {"type": "string"},
@@ -220,14 +219,23 @@ class AIPlanner:
                     },
                     "required": ["actions"],
                 },
-                options={"temperature": 0.3, "num_ctx": 8192},
+                "think": False,
+                "stream": False,
+                "options": {"temperature": 0.3, "num_ctx": 8192},
+            }
+            resp = httpx.post(
+                f"{OLLAMA_HOST}/api/chat",
+                json=payload,
+                timeout=300.0,
             )
+            resp.raise_for_status()
+            data = resp.json()
         finally:
             stop_event.set()
             spinner.join()
 
         dt = time.perf_counter() - t0
-        content = response.message.content
+        content = data["message"]["content"]
         print(f"[AI] 응답 수신 ({dt:.2f}s, {len(content)}자)")
         if _debug_enabled():
             print(f"[AI][디버그] raw={content[:300]}")