docs/diagrams/system.md

```mermaid
graph TD
    A[Client] -->|HTTP/WebSocket| B[FastAPI Server]
    B -->|Authentication| C[Auth Module]
    B -->|Text Generation| D[Generation Module]
    B -->|Rate Limiting| E[Rate Limiter]
    B -->|Logging| F[Logging Middleware]
    C -->|JWT Token| G[Secrets Manager]
    D -->|Model Inference| H[Hugging Face Model]
    E -->|Rate Limit Check| I[Redis/Database]
    F -->|Logs| J[Log Storage]
    H -->|Streaming| A

    subgraph FastAPI Server
        B
        C
        D
        E
        F
    end

    subgraph External Services
        G[Secrets Manager]
        H[Hugging Face Model]
        I[Redis/Database]
        J[Log Storage]
    end
```