BenBox

Name: BenBox
Author: DrBenjamin

@DrBenjamin

代码仓库

关于 BenBox

Agent AI app utilizing MCP tools with Angular mobile and Phoenix desktop app.

#python #angular #phoenix #js #ts #image-recognition #streamlit #mcp-server

基本信息

分类

其他

运行时

node

传输方式

stdio

发布者

DrBenjamin

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "BenBox": {
      "command": "python",
      "args": [
        "-m",
        "pip",
        "install",
        "mcp[cli]"
      ]
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is BenBox?

BenBox is an Agent AI application that uses Model Context Protocol (MCP) tools, integrating with Ollama, Azure OpenAI, Snowflake, PostgreSQL, and MinIO. It provides an Angular mobile app and a Phoenix desktop app, targeting developers building cloud-native AI agents for image recognition and file organization.

How to use BenBox?

Clone the repository, create a conda environment from environment.yml, install the mcp package, and run the MCP server (python src/server.py or mcp dev src/server.py). Launch the Streamlit app with streamlit run app.py, the Angular mobile app with ng serve, or the desktop app with python BenBox.py. Configure secrets in .streamlit/st.secrets.toml (e.g., MCP_URL, AZURE_OPENAI_API_KEY, Ollama settings).

Key features of BenBox

SSE‑based MCP server for decoupled agent communication.
Supports Ollama and Azure OpenAI as LLM providers.
Image recognition via Streamlit upload or MCP Inspector.
Angular mobile app with Capacitor for iOS.
Desktop app (PyInstaller) for file organization.
Docker support for containerized deployment.
Integration with Snowflake, PostgreSQL, and MinIO storage.

Use cases of BenBox

Build an AI agent that recognizes images using a local or cloud LLM.
Run a mobile image‑upload interface connected to an MCP server.
Organize and manage files with a desktop GUI backed by MinIO.
Deploy a cloud‑native agent with decoupled MCP client and server.

FAQ from BenBox

What runtime dependencies does BenBox require?

BenBox requires Python 3, Node.js, npm, conda, Ollama (for local models), and optionally Azure CLI for Azure OpenAI.

Where does data persist in BenBox?

User data can be stored in Snowflake or PostgreSQL, and file storage uses MinIO buckets. Configuration is set via .streamlit/st.secrets.toml.

How does the MCP server communicate?

The MCP server runs on HTTP (SSE) — configurable via MCP_URL in secrets, defaulting to http://127.0.0.1:8080. Clients connect over this transport.

How is authentication handled?

Azure OpenAI requires az login (or az login --use-device-code). MinIO uses access/secret keys. No built‑in auth for the MCP server itself is documented.

Can BenBox run in a container?

Yes, Dockerfiles are provided for the MCP server (Dockerfile_MCP), Streamlit app, Angular app, and the desktop VNC container. Use docker-compose up -d to start all services.