feat: оновлення інфраструктури з Node #2 та нові сервіси

2025-11-21 00:35:03 -08:00
parent 36c35cab57
commit 31f3602047
5 changed files with 179 additions and 18 deletions
--- a/INFRASTRUCTURE.md
+++ b/INFRASTRUCTURE.md
@@ -6,20 +6,36 @@

 ---

-## 📍 Production Servers
+## 📍 Network Nodes

-### GEX44 Server #2844465 (Hetzner)
+### Node #1: Production Server (Hetzner GEX44 #2844465)
+- **Node ID:** `node-1-hetzner-gex44`
 - **IP Address:** `144.76.224.179`
 - **SSH Access:** `ssh root@144.76.224.179`
- **Location:** Hetzner Cloud
- **Project Root:** `/opt/microdao-daarion` (or `/opt/dagi-router`)
+- **Location:** Hetzner Cloud (Germany)
+- **Project Root:** `/opt/microdao-daarion`
 - **Docker Network:** `dagi-network`
+- **Role:** Production Router + Gateway + All Services
+- **Uptime:** 24/7

 **Domains:**
 - `gateway.daarion.city` → `144.76.224.179` (Gateway + Nginx)
 - `api.daarion.city` → TBD (API Gateway)
 - `daarion.city` → TBD (Main website)

+### Node #2: Development Node (MacBook Pro M4 Max)
+- **Node ID:** `node-2-macbook-m4max`
+- **Local IP:** `192.168.1.244`
+- **SSH Access:** `ssh apple@192.168.1.244` (if enabled)
+- **Location:** Local Network (Ivan's Office)
+- **Project Root:** `/Users/apple/github-projects/microdao-daarion`
+- **Role:** Development + Testing + Backup Router
+- **Specs:** M4 Max (16 cores), 64GB RAM, 2TB SSD, 40-core GPU
+- **Uptime:** On-demand (battery-powered)
+
+**See full specs:** [NODE-2-MACBOOK-SPECS.md](./NODE-2-MACBOOK-SPECS.md)  
+**Current state:** [NODE-2-CURRENT-STATE.md](./NODE-2-CURRENT-STATE.md) — What's running now
+
 ---

 ## 🐙 GitHub Repositories
--- a/docs/infrastructure_quick_ref.ipynb
+++ b/docs/infrastructure_quick_ref.ipynb
@@ -6,12 +6,12 @@
   "source": [
    "# 🚀 Infrastructure Quick Reference — DAARION & MicroDAO\n",
    "\n",
-    "**Версія:** 1.1.0  \n",
+    "**Версія:** 1.2.0  \n",
    "**Останнє оновлення:** 2025-01-17  \n",
    "\n",
    "Цей notebook містить швидкий довідник по серверах, репозиторіях та endpoints для DAGI Stack.\n",
    "\n",
-    "**NEW:** Vision Encoder + Qdrant vector database (OpenCLIP ViT-L/14)"
+    "**NEW:** Vision Encoder + Qdrant + Node #2 (MacBook Pro M4 Max)"
   ]
  },
  {
@@ -48,6 +48,75 @@
    "    print(f\"{name.upper():<20} {service['port']:<7} {service['container']:<30} {health}{gpu}\")"
   ]
  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 🖥️ Network Nodes\n",
+    "\n",
+    "### Node #1: Production Server (Hetzner)\n",
+    "- **Node ID:** node-1-hetzner-gex44\n",
+    "- **IP:** 144.76.224.179\n",
+    "- **Role:** Production Router + Gateway + All Services (24/7)\n",
+    "- **Location:** Hetzner Cloud (Germany)\n",
+    "\n",
+    "### Node #2: Development Node (MacBook Pro M4 Max)\n",
+    "- **Node ID:** node-2-macbook-m4max\n",
+    "- **Local IP:** 192.168.1.244\n",
+    "- **Role:** Development + Testing + Backup Router\n",
+    "- **Specs:** M4 Max (16 cores), 64GB RAM, 2TB SSD, 40-core GPU\n",
+    "- **Location:** Local Network (Ivan's Office)\n",
+    "- **Docs:** [NODE-2-MACBOOK-SPECS.md](../NODE-2-MACBOOK-SPECS.md)\n",
+    "\n",
+    "---"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Network Nodes Configuration\n",
+    "NODES = {\n",
+    "    \"node-1\": {\n",
+    "        \"name\": \"Hetzner GEX44\",\n",
+    "        \"ip\": \"144.76.224.179\",\n",
+    "        \"local_ip\": None,\n",
+    "        \"role\": \"production\",\n",
+    "        \"uptime\": \"24/7\",\n",
+    "        \"ssh\": \"root@144.76.224.179\",\n",
+    "        \"domain\": \"gateway.daarion.city\",\n",
+    "        \"services\": \"All (17 services)\",\n",
+    "        \"specs\": \"See SYSTEM-INVENTORY.md\"\n",
+    "    },\n",
+    "    \"node-2\": {\n",
+    "        \"name\": \"MacBook Pro M4 Max\",\n",
+    "        \"ip\": None,\n",
+    "        \"local_ip\": \"192.168.1.244\",\n",
+    "        \"role\": \"development\",\n",
+    "        \"uptime\": \"on-demand\",\n",
+    "        \"ssh\": \"apple@192.168.1.244\",\n",
+    "        \"domain\": None,\n",
+    "        \"services\": \"Core only (Router, DevTools, Memory, Ollama)\",\n",
+    "        \"specs\": \"M4 Max, 16 cores, 64GB RAM, 2TB SSD, 40-core GPU\"\n",
+    "    }\n",
+    "}\n",
+    "\n",
+    "print(\"DAGI Stack Network Nodes:\")\n",
+    "print(\"=\"*80)\n",
+    "for node_id, node in NODES.items():\n",
+    "    print(f\"\\n{node_id.upper()}: {node['name']}\")\n",
+    "    print(f\"  Role:     {node['role']}\")\n",
+    "    print(f\"  IP:       {node['ip'] or node['local_ip']}\")\n",
+    "    print(f\"  SSH:      {node['ssh']}\")\n",
+    "    print(f\"  Uptime:   {node['uptime']}\")\n",
+    "    print(f\"  Services: {node['services']}\")\n",
+    "    if node['domain']:\n",
+    "        print(f\"  Domain:   https://{node['domain']}\")\n",
+    "    print(f\"  Specs:    {node['specs']}\")"
+   ]
+  },
  {
   "cell_type": "markdown",
   "metadata": {},
@@ -157,7 +226,8 @@
    "    \"Deployment\": \"../DEPLOY-NOW.md\",\n",
    "    \"Helion Status\": \"../STATUS-HELION.md\",\n",
    "    \"Architecture Index\": \"../docs/cursor/README.md\",\n",
-    "    \"API Reference\": \"../docs/api.md\"\n",
+    "    \"API Reference\": \"../docs/api.md\",\n",
+    "    \"Node #2 Specs\": \"../NODE-2-MACBOOK-SPECS.md\"\n",
    "}\n",
    "\n",
    "print(\"Documentation Quick Links:\")\n",
@@ -180,11 +250,14 @@
    "- ✅ **768-dim embeddings** for multimodal RAG\n",
    "- ✅ Created VISION-ENCODER-STATUS.md with full implementation details\n",
    "- ✅ Added test-vision-encoder.sh smoke tests\n",
+    "- ✅ **Added Node #2** (MacBook Pro M4 Max) as development node\n",
+    "- ✅ Created NODE-2-MACBOOK-SPECS.md with complete specifications\n",
    "\n",
-    "### Services Count: 17 (from 15)\n",
-    "- Total Services: 17\n",
-    "- GPU Services: 1 (Vision Encoder)\n",
-    "- Vector Databases: 1 (Qdrant)\n",
+    "### Network Architecture\n",
+    "- **Nodes:** 2 (1 production + 1 development)\n",
+    "- **Total Services:** 17\n",
+    "- **GPU Services:** 1 (Vision Encoder) + Node #2 (40-core Apple GPU)\n",
+    "- **Vector Databases:** 1 (Qdrant)\n",
    "\n",
    "---\n",
    "\n",
--- a/http_api.py
+++ b/http_api.py
@@ -5,7 +5,7 @@ Provides HTTP endpoints for routing requests
 """

 from fastapi import APIRouter, HTTPException, status
-from pydantic import BaseModel, Field
+from pydantic import BaseModel, Field, ConfigDict
 from typing import Optional, Dict, Any
 import logging

@@ -21,6 +21,8 @@ logger = logging.getLogger(__name__)

 class IncomingRequest(BaseModel):
    """HTTP request format"""
+    model_config = ConfigDict(extra='allow')  # Дозволити додаткові поля для сумісності
+    
    mode: Optional[str] = Field(None, description="Request mode (e.g., 'chat', 'crew')")
    agent: Optional[str] = Field(None, description="Agent ID (e.g., 'devtools')")
    message: Optional[str] = Field(None, description="User message")
@@ -29,6 +31,7 @@ class IncomingRequest(BaseModel):
    session_id: Optional[str] = Field(None, description="Session identifier")
    user_id: Optional[str] = Field(None, description="User identifier")
    payload: Dict[str, Any] = Field(default_factory=dict, description="Additional payload data")
+    context: Optional[Dict[str, Any]] = Field(None, description="Legacy: context on top level (will be migrated to payload.context)")


 class RouterAPIResponse(BaseModel):
@@ -74,6 +77,32 @@ def build_router_http(app_core: RouterApp) -> APIRouter:
        - Other metadata
        """
        logger.info(f"Incoming request: agent={req.agent}, mode={req.mode}")
+        logger.info(f"Raw payload type: {type(req.payload)}, keys: {list(req.payload.keys()) if req.payload else []}")
+        logger.info(f"Raw context: {req.context}")
+        
+        # Нормалізувати payload: якщо є "context" на верхньому рівні (старий формат),
+        # перемістити його в payload.context для уніфікованої обробки
+        normalized_payload = req.payload.copy() if req.payload else {}
+        
+        # Перевірити, чи є context на верхньому рівні (legacy формат від gateway-bot)
+        # Це потрібно для сумісності з DAARWIZZ
+        if req.context:
+            # Якщо context на верхньому рівні, перемістити в payload.context
+            if "context" not in normalized_payload:
+                normalized_payload["context"] = {}
+            # Мержити context з верхнього рівня в payload.context
+            if isinstance(req.context, dict):
+                normalized_payload["context"].update(req.context)
+                logger.info(f"✅ Migrated top-level context to payload.context for agent={req.agent}, keys={list(req.context.keys())}")
+        
+        logger.info(f"✅ Normalized payload keys: {list(normalized_payload.keys())}")
+        if normalized_payload and "context" in normalized_payload:
+            logger.info(f"✅ Context keys: {list(normalized_payload['context'].keys()) if isinstance(normalized_payload.get('context'), dict) else []}")
+            if isinstance(normalized_payload.get('context'), dict) and "system_prompt" in normalized_payload['context']:
+                sp = normalized_payload['context']['system_prompt']
+                sp_len = len(sp) if sp else 0
+                logger.info(f"✅ System prompt found in context: {sp_len} chars")
+                logger.info(f"✅ System prompt preview: {sp[:100] if sp else 'None'}...")
        
        # Convert to internal RouterRequest
        rreq = RouterRequest(
@@ -84,7 +113,7 @@ def build_router_http(app_core: RouterApp) -> APIRouter:
            session_id=req.session_id,
            user_id=req.user_id,
            message=req.message,
-            payload=req.payload,
+            payload=normalized_payload,
        )
        
        # Handle request
--- a/providers/llm_provider.py
+++ b/providers/llm_provider.py
@@ -158,15 +158,34 @@ class LLMProvider(Provider):
    def _get_system_prompt(self, req: RouterRequest) -> Optional[str]:
        """Get system prompt based on agent or context"""
        # 1. Check if context.system_prompt provided (e.g., from Gateway)
+        logger.info(f"[DEBUG] _get_system_prompt called for agent={req.agent}")
+        logger.info(f"[DEBUG] req.payload type: {type(req.payload)}, keys: {list(req.payload.keys()) if req.payload else []}")
        context = req.payload.get("context") or {}
-        logger.info(f"[DEBUG] payload keys: {list(req.payload.keys())}")
-        logger.info(f"[DEBUG] context keys: {list(context.keys())}")
-        if "system_prompt" in context:
+        logger.info(f"[DEBUG] context type: {type(context)}, keys: {list(context.keys()) if isinstance(context, dict) else 'not a dict'}")
+        if isinstance(context, dict) and "system_prompt" in context:
            prompt = context["system_prompt"]
-            logger.info(f"[DEBUG] Using context.system_prompt: {len(prompt)} chars, agent={req.agent}")
+            logger.info(f"[DEBUG] ✅ Using context.system_prompt: {len(prompt)} chars, agent={req.agent}")
+            logger.info(f"[DEBUG] System prompt type: {type(prompt)}")
+            logger.info(f"[DEBUG] System prompt preview (first 200): {str(prompt)[:200]}...")
+            logger.info(f"[DEBUG] System prompt full length check: {len(str(prompt))} chars")
+            # Переконаємось, що це рядок
+            if not isinstance(prompt, str):
+                logger.warning(f"[DEBUG] ⚠️ System prompt is not a string! Type: {type(prompt)}, value: {prompt}")
+                prompt = str(prompt) if prompt else None
            return prompt
+        else:
+            logger.info(f"[DEBUG] ⚠️ No system_prompt in context for agent={req.agent}")
+        
+        # 2. Agent-specific system prompts (fallback, якщо не передано в context)
+        if req.agent == "helion":
+            return (
+                "Ти - Helion, AI-агент платформи Energy Union екосистеми DAARION.city. "
+                "Допомагай користувачам з технологіями EcoMiner/BioMiner, токеномікою та DAO governance. "
+                "Твої основні функції: консультації з енергетичними технологіями, пояснення токеноміки Energy Union, "
+                "допомога з onboarding в DAO, відповіді на питання про EcoMiner/BioMiner устаткування. "
+                "Стиль спілкування: професійний, технічний, але зрозумілий, точний у цифрах та даних."
+            )
        
-        # 2. Agent-specific system prompts
        if req.agent == "daarwizz":
            return (
                "Ти — DAARWIZZ, офіційний AI-агент екосистеми DAARION.city. "
--- a/router-config.yml
+++ b/router-config.yml
@@ -89,6 +89,30 @@ agents:
      You are a multi-agent orchestrator for DAARION.city microDAO ecosystem.
      You coordinate complex workflows involving multiple specialized agents.
  
+  greenfood:
+    description: "GREENFOOD Assistant - ERP orchestrator for craft food producers"
+    default_llm: local_qwen3_8b
+    system_prompt: |
+      Ти — GREENFOOD Assistant, фронтовий оркестратор ERP-системи для крафтових виробників, хабів та покупців.
+      Твоя місія: зрозуміти, хто з тобою говорить (комітент, менеджер складу, логіст, бухгалтер, маркетолог, покупець), 
+      виявити намір і делегувати завдання спеціалізованим агентам GREENFOOD.
+      
+      У твоєму розпорядженні 12 спеціалізованих агентів:
+      - Product & Catalog (каталог товарів)
+      - Batch & Quality (партії та якість)
+      - Vendor Success (успіх комітентів)
+      - Warehouse (склад)
+      - Logistics & Delivery (доставка)
+      - Seller (продажі)
+      - Customer Care (підтримка)
+      - Finance & Pricing (фінанси)
+      - SMM & Campaigns (маркетинг)
+      - SEO & Web (SEO)
+      - Analytics & BI (аналітика)
+      - Compliance & Audit (аудит)
+      
+      Відповідай українською, чітко та по-діло вому.
+  
  helion:
    description: "Helion - AI agent for Energy Union platform"
    default_llm: local_qwen3_8b