Update README.md

README.md

@@ -106,39 +106,12 @@ RewardPilot is an autonomous AI agent that analyzes your transaction in real-tim
 
 ---
 
+
 ## 🏗️ Architecture
 
 ### Multi-Agent MCP System
 
-```mermaid
-graph TB
-    User[👤 User Input] --> Gradio[🎨 Gradio Interface]
-    Gradio --> Orchestrator[🎯 Orchestrator<br/>Claude 3.5 Sonnet]
-    
-    Orchestrator --> Phase1[📋 Phase 1: Planning<br/>Analyze context + strategy]
-    Phase1 --> Phase2[⚡ Phase 2: Execution<br/>Parallel MCP calls]
-    Phase2 --> Phase3[🧠 Phase 3: Reasoning<br/>Gemini 2.0 synthesis]
-    Phase3 --> Phase4[📚 Phase 4: Response<br/>Format + deliver]
-    
-    Phase2 --> MCP1[🔧 Smart Wallet MCP<br/>Card recommendations]
-    Phase2 --> MCP2[📚 Rewards RAG MCP<br/>Semantic search]
-    Phase2 --> MCP3[📈 Forecast MCP<br/>Spending predictions]
-    
-    MCP1 --> DB1[(💳 Card Database<br/>50+ cards)]
-    MCP2 --> Vector[(🔍 Vector Store<br/>LlamaIndex)]
-    MCP3 --> ML[(🤖 ML Models<br/>Forecasting)]
-    
-    Phase3 --> Gemini[🎯 Gemini 2.0 Flash<br/>Context synthesis]
-    
-    Phase4 --> Output[✅ Recommendation<br/>+ Explanation]
-    Output --> Gradio
-    
-    style Orchestrator fill:#667eea,color:#fff
-    style Gemini fill:#4285F4,color:#fff
-    style MCP1 fill:#10b981,color:#fff
-    style MCP2 fill:#8b5cf6,color:#fff
-    style MCP3 fill:#f59e0b,color:#fff
-```
+![Architecture Diagram](https://huggingface.co/spaces/MCP-1st-Birthday/rewardpilot-web-ui/blob/main/mermaid.png)
 
 ### MCP Microservices