Skip to main content
Glama
DamilolaAdegunwa

MCP Sandboxed Python Data Analyst

by DamilolaAdegunwa
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

MCP Sandboxed Python Data Analyst

CI Pipeline

Google ADK Master Curriculum — Track 3: MCP (Model Context Protocol) & Tool Ecosystems (Application 15 of 40)

Architectural Overview

Secure MCP data analyst executing statistical Python code in isolated micro-VM sandboxes, producing charts, and returning analytical summaries.

Google ADK Primitives Featured

  • Agent

  • bash_tool

  • mcp_tool

Related MCP server: Code Executor MCP Server

System Architecture

graph TD
    User["Client / Event Stream"] --> Gateway["ADK Runtime / FastApi Server"]
    Gateway --> Agent["adk-mcp-sandboxed-data-analyst (Root Agent)"]
    Agent --> Tools["Specialized Domain Tools"]
    Tools --> External["External Systems / Google Cloud APIs"]

Project Structure

.
├── agent.py              # Google ADK agent definition
├── config.py             # Environment & model configurations
├── tools/                # Specialized domain tools
├── tests/                # Automated pytest suite
├── main.py               # Application entrypoint
├── Dockerfile            # Container deployment
└── .github/workflows/    # Automated CI verification

Running Locally

# 1. Install dependencies
pip install -r requirements.txt

# 2. Configure credentials
export GOOGLE_API_KEY="your-gemini-api-key"

# 3. Run application
python main.py

Running Tests

pytest tests/

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Resources

Related MCP Connectors

Related MCP Servers

  • Netmind Code Interpreter

    protagolabs
    A
    license
    Not graded
    quality
    D
    maintenance
    Enables secure cloud-based execution of code across 14+ programming languages within a sandboxed environment. It supports file management, standard input/output handling, and automatic generation of visual artifacts like plots and charts.
    MIT

  • Code Executor MCP Server

    CSOAI-ORG
    A
    license
    Not graded
    quality
    A
    maintenance
    Provides sandboxed code execution for AI agents with support for Python, JavaScript, and shell commands. Includes comprehensive safety features like destructive pattern blocking, timeout protection, and restricted file access for secure production use.
    12 npm
    113 PyPI
    MIT

  • mcp-run-isolated-python

    Kigstn
    A
    license
    Not graded
    quality
    C
    maintenance
    Enables LLMs to execute Python code securely in a sandboxed environment. Supports configurable restrictions like no network access and returns results including files.
    MIT