fix(tarot): Claude 응답 시간 단축 + nginx timeout 정리

504 Gateway Timeout 근본 원인은 DSM Reverse Proxy의 60s 기본 timeout
(agent-office는 200 OK 정상 응답했으나 client 도달 전 DSM이 끊음).
사용자 측 DSM Reverse Proxy timeout 늘리기 별도 필요.

코드 측 대응:
- pipeline.py max_tokens 2048 → 1400 (응답 시간 단축, 3-card spread 충분)
- pipeline.py에 latency_ms·tokens 로그 출력 (모니터링)
- nginx /api/agent-office/에 proxy_send_timeout 300s, proxy_connect_timeout 60s
  추가 (proxy_read_timeout은 WebSocket 위해 86400s 유지)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

agent-office/app/tarot/pipeline.py

@@ -1,5 +1,6 @@
 """Tarot 파이프라인 — Claude Sonnet 호출 + 파싱 폴백 + reroll 1회."""
 import json
+import logging
 import time
 from typing import Any, Dict
 
@@ -12,6 +13,9 @@ from ..config import (
     TAROT_COST_OUTPUT_PER_M,
     TAROT_TIMEOUT_SEC,
 )
+
+
+logger = logging.getLogger("agent-office.tarot")
 from ..models import TarotInterpretRequest
 from .prompt import SYSTEM_PROMPT, build_user_message
 from .schema import validate_interpretation
@@ -62,7 +66,7 @@ async def _call_claude(user_text: str, feedback: str = "") -> tuple[dict, dict,
         user_text = f"이전 응답이 다음 이유로 거절됨: {feedback}\n올바른 스키마(시스템 지침)로 다시 응답.\n\n{user_text}"
     payload = {
         "model": TAROT_MODEL,
-        "max_tokens": 2048,
+        "max_tokens": 1400,  # 응답 시간 단축 — 3-card spread evidence·interactions 포함 충분
         "system": [{"type": "text", "text": SYSTEM_PROMPT,
                     "cache_control": {"type": "ephemeral"}}],
         "messages": [{"role": "user", "content": [{"type": "text", "text": user_text}]}],
@@ -82,11 +86,14 @@ async def _call_claude(user_text: str, feedback: str = "") -> tuple[dict, dict,
     raw_text = "".join(
         b.get("text", "") for b in resp.get("content", []) if b.get("type") == "text"
     )
-    parsed = _extract_json(raw_text)
     usage = resp.get("usage", {}) or {}
+    tokens_in = int(usage.get("input_tokens", 0) or 0)
+    tokens_out = int(usage.get("output_tokens", 0) or 0)
+    logger.info("tarot claude call: latency=%dms, in=%d, out=%d", latency_ms, tokens_in, tokens_out)
+    parsed = _extract_json(raw_text)
     meta = {
-        "tokens_in": int(usage.get("input_tokens", 0) or 0),
+        "tokens_in": tokens_in,
-        "tokens_out": int(usage.get("output_tokens", 0) or 0),
+        "tokens_out": tokens_out,
         "latency_ms": latency_ms,
     }
     return parsed, meta, raw_text

nginx/default.conf

@@ -357,6 +357,8 @@ server {
     proxy_set_header Upgrade $http_upgrade;
     proxy_set_header Connection "upgrade";
     proxy_read_timeout 86400s;
+    proxy_send_timeout 300s;
+    proxy_connect_timeout 60s;
     proxy_pass http://$agent_office_backend$request_uri;
   }