BenBox
About
Agent AI app utilizing MCP tools with Angular mobile and Phoenix desktop app.
Explore
- 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.
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
BenBoxCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Install the required packages and the MCP server and client:
sudo apt install nodejs npm cocoapods
brew install nodejs npm cocoapods
conda install -c conda-forge mcp
conda env create -f environment.yml
conda activate benbox
python -m pip install "mcp[cli]"
mcp dev src/server.py
python src/server.py
python -m streamlit run app.py
To configure the Azure OpenAI API, you need to install the Azure CLI and
Azure Dev CLI. Use the following commands to install them:
curl -fsSL https://aka.ms/install-azd.sh | bash
To allow public (anonymous) access to a specific bucket, use the following command:
[LLM]
LLM_CHATBOT_NAME = "<chatbot_name>"
LLM_SYSTEM = "Please write a short answer."
LLM_SYSTEM_PLUS = "Prioritize the most relevant information from the similarity search!"
LLM_ASSISTANT = "How can I help?"
LLM_USER_EXAMPLE = "<user_example>"
LLM_ASSISTANT_EXAMPLE = "<>assistant_example>"
[snowflake]
user = "<user_name>"
account = "<account_name>"
private_key_file = "<path to rsa_key.p8>"
role = "<role_name>"
warehouse = "<warehouse_name>"
database = "<database_name>"
schema = "<schema_name>"
[psotgresql]
user = "<user_name>"
password = "<password>"
host = "<host_name>"
port = "<port_number>"
database = "<database_name>"
table = "<table_name>"
[MinIO]
endpoint = "http://127.0.0.1:9000"
bucket = "<bucket_name>"
access_key = "<access_key>"
secret_key = "<secret_key>"
systemctl start ollama. # Linux
brew services start ollama # or Mac
ollama run llama3.2-vision
python -m pip install dbt-core dbt-snowflake
dbt deps
dbt run
dbt test
pytest -v --tb=short --disable-warnings --maxfail=1
npm install -g @angular/cli
npm install
ng serve
The app will be available at http://localhost:4200.
brew install minio/stable/mc
python BenBox.py
```
docker-compose down
docker run -it --rm -p 6080:6080 benbox-vnc
docker run -it --rm -p 8501:8501 streamlit
docker run -it --rm -p 8080:8080 mcp
docker-compose up -d
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"benbox": {
"BenBox": {
"command": "python",
"args": [
"-m",
"pip",
"install",
"mcp[cli]"
]
}
}
}
}McpServers
{
"BenBox": {
"command": "python",
"args": [
"-m",
"pip",
"install",
"mcp[cli]"
]
}
}BenBox is an Agent AI app utilizing MCP
tools with Angular mobile and Phoenix desktop app.
Why MCP?
MCP server can now be some running process that agents (clients) connect to,
use, and disconnect from whenever and wherever they want. In other words,
an SSE-based server and clients can be decoupled processes
(potentially even, on decoupled nodes). This is different and better fits
"cloud-native" use-cases compared to the STDIO-based pattern where the client
itself spawns the server as a subprocess.
Setup
Install the required packages and the MCP server and client:
```bash
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
