feat(production): sync all modified production files to git

Includes updates across gateway, router, node-worker, memory-service,
aurora-service, swapper, sofiia-console UI and node2 infrastructure:

- gateway-bot: Dockerfile, http_api.py, druid/aistalk prompts, doc_service
- services/router: main.py, router-config.yml, fabric_metrics, memory_retrieval,
  offload_client, prompt_builder
- services/node-worker: worker.py, main.py, config.py, fabric_metrics
- services/memory-service: Dockerfile, database.py, main.py, requirements
- services/aurora-service: main.py (+399), kling.py, quality_report.py
- services/swapper-service: main.py, swapper_config_node2.yaml
- services/sofiia-console: static/index.html (console UI update)
- config: agent_registry, crewai_agents/teams, router_agents
- ops/fabric_preflight.sh: updated preflight checks
- router-config.yml, docker-compose.node2.yml: infra updates
- docs: NODA1-AGENT-ARCHITECTURE, fabric_contract updated

Made-with: Cursor

services/router/memory_retrieval.py

@@ -20,6 +20,7 @@ import json
 import logging
 import re
 import hashlib
+from time import monotonic
 from typing import Optional, Dict, Any, List
 from dataclasses import dataclass, field
 from datetime import datetime
@@ -41,6 +42,20 @@ PENDING_QUESTIONS_LIMIT = int(os.getenv("AGENT_PENDING_QUESTIONS_LIMIT", "5"))
 SHARED_AGRO_LIBRARY_ENABLED = os.getenv("AGROMATRIX_SHARED_LIBRARY_ENABLED", "true").lower() == "true"
 SHARED_AGRO_LIBRARY_REQUIRE_REVIEW = os.getenv("AGROMATRIX_SHARED_LIBRARY_REQUIRE_REVIEW", "true").lower() == "true"
 DOC_VERSION_PREVIEW_CHARS = int(os.getenv("DOC_VERSION_PREVIEW_CHARS", "240"))
+WARNING_THROTTLE_SECONDS = float(os.getenv("MEMORY_RETRIEVAL_WARNING_THROTTLE_S", "60") or "60")
+_warning_last_ts: Dict[str, float] = {}
+
+
+def _warning_throttled(key: str, message: str) -> None:
+    """Emit repetitive warnings at most once per throttle window."""
+    if WARNING_THROTTLE_SECONDS <= 0:
+        logger.warning(message)
+        return
+    now = monotonic()
+    last = _warning_last_ts.get(key, 0.0)
+    if now - last >= WARNING_THROTTLE_SECONDS:
+        _warning_last_ts[key] = now
+        logger.warning(message)
 
 
 @dataclass
@@ -1067,7 +1082,7 @@ class MemoryRetrieval:
                 )
             return True
         except Exception as e:
-            logger.warning(f"register_pending_question failed: {e}")
+            _warning_throttled("register_pending_question_failed", f"register_pending_question failed: {e}")
             return False
 
     async def resolve_pending_question(
@@ -1086,7 +1101,7 @@ class MemoryRetrieval:
                 row = await conn.fetchrow(
                     """
                     WITH target AS (
-                      SELECT id
+                      SELECT id, question_fingerprint
                       FROM agent_pending_questions
                       WHERE channel = $1
                         AND chat_id = $2
@@ -1095,17 +1110,49 @@ class MemoryRetrieval:
                         AND status = 'pending'
                       ORDER BY created_at ASC
                       LIMIT 1
+                    ), decision AS (
+                      SELECT
+                        t.id,
+                        CASE
+                          WHEN $5 = 'dismissed' THEN 'dismissed'
+                          WHEN EXISTS (
+                            SELECT 1
+                            FROM agent_pending_questions q
+                            WHERE q.channel = $1
+                              AND q.chat_id = $2
+                              AND q.user_id = $3
+                              AND q.agent_id = $4
+                              AND q.status = 'answered'
+                              AND q.question_fingerprint = t.question_fingerprint
+                          ) THEN 'dismissed'
+                          ELSE 'answered'
+                        END AS next_status,
+                        CASE
+                          WHEN $5 = 'dismissed' THEN $5
+                          WHEN EXISTS (
+                            SELECT 1
+                            FROM agent_pending_questions q
+                            WHERE q.channel = $1
+                              AND q.chat_id = $2
+                              AND q.user_id = $3
+                              AND q.agent_id = $4
+                              AND q.status = 'answered'
+                              AND q.question_fingerprint = t.question_fingerprint
+                          ) THEN 'duplicate_answered'
+                          ELSE $5
+                        END AS resolution_reason
+                      FROM target t
                     )
                     UPDATE agent_pending_questions p
-                    SET status = CASE WHEN $5 = 'dismissed' THEN 'dismissed' ELSE 'answered' END,
+                    SET status = d.next_status,
                         answered_at = NOW(),
                         metadata = COALESCE(p.metadata, '{}'::jsonb)
                                    || jsonb_build_object(
-                                        'resolution_reason', $5,
+                                        'resolution_reason', d.resolution_reason,
                                         'answer_fingerprint', COALESCE($6, '')
                                       )
-                    FROM target t
-                    WHERE p.id = t.id
+                    FROM decision d
+                    WHERE p.id = d.id
                     RETURNING p.id
                     """,
                     channel,
@@ -1117,7 +1164,7 @@ class MemoryRetrieval:
                 )
             return bool(row)
         except Exception as e:
-            logger.warning(f"resolve_pending_question failed: {e}")
+            _warning_throttled("resolve_pending_question_failed", f"resolve_pending_question failed: {e}")
             return False
 
     @staticmethod