A multi-layered cloud-native architecture for a development platform, featuring browser access, control plane, runtime orchestration, per-user runtimes, an
flowchart TB
A([Browser])
M([本地文件])
subgraph L1[入口层]
direction TB
B[Traefik]
end
subgraph L2[平台层]
direction TB
D[ClawLoops 控制平面]
E[Web UI]
K[文件管理服务]
N[文件上传服务]
end
subgraph L3[编排层]
direction TB
F[Runtime Manager]
end
subgraph L4[运行时层]
direction TB
G[每用户 OpenClaw Runtime]
L[容器文件系统]
end
subgraph L5[模型层]
direction TB
H[LiteLLM]
I[上游模型提供商]
end
J[(数据存储)]
A -->|访问入口| B -->|路由转发| D
B -->|静态资源| E
E -->|API 调用| D
E -->|文件管理操作| K
E -->|文件上传| N
M -->|选择文件| E
K -->|文件操作请求| D
N -->|上传请求| D
D -->|生命周期管理| F -->|启动/停止/管理| G
D -->|文件操作| F -->|容器文件操作| L
D -->|文件上传| F -->|文件写入| L
D -->|模型访问| H -->|代理请求| I
D -->|数据持久化| J
classDef edge fill:#eff6ff,stroke:#2563eb,stroke-width:2px,color:#0f172a;
classDef plane fill:#dcfce7,stroke:#16a34a,stroke-width:2px,color:#0f172a;
classDef runtime fill:#f3e8ff,stroke:#7c3aed,stroke-width:2px,color:#0f172a;
classDef model fill:#ffe4e6,stroke:#db2777,stroke-width:2px,color:#0f172a;
classDef data fill:#f8fafc,stroke:#64748b,stroke-width:2px,color:#0f172a;
classDef file fill:#fef3c7,stroke:#d97706,stroke-width:2px,color:#0f172a;
class A,B edge;
class D,E,K,N plane;
class F,G,L runtime;
class H,I model;
class J data;
class K,L,M,N file;This diagram illustrates the five-layer architecture of the ClawLoops platform: the Entry Layer (Traefik) for routing and authentication, the Platform Layer (Control Plane, Web UI, File Management) for core logic, the Orchestration Layer (Runtime Manager) for managing user runtimes, the Runtime Layer (OpenClaw Runtime, Container File System) providing isolated environments, and the Model Layer (LiteLLM, Upstream Models) for AI model access. It also shows interactions with local files and data storage.
Use this diagram to understand the architectural breakdown of a complex cloud-native development platform, a microservices-based system with distinct functional layers, or a system requiring secure, isolated per-user runtimes and AI model integration.
This architecture can be adapted by swapping out specific components (e.g., different ingress controllers, alternative container orchestration, different file storage solutions, or various LLM providers). Additional services can be integrated into existing layers, or new layers can be introduced for specific functionalities like monitoring, logging, or advanced security.