MCP Server Template
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
MCP Server Template
A scaffold template for building MCP (Model Context Protocol) servers using the oxsci-oma-mcp framework.
Prerequisites
Python 3.11 or higher
Poetry for dependency management
AWS CLI configured with access to CodeArtifact (for installing dependencies from CodeArtifact)
Note: Docker is only required for CI/CD deployment workflows, not for local development.
Installation
Installation Prerequisites
The installer script itself only requires Python 3.11+ (uses only Python standard library, no external packages needed).
However, to complete the installation, you also need:
Git installed (for initializing the repository)
Network connectivity to GitHub (for downloading the template)
Write permissions in the directory where you'll run the script
The installer will also check for:
AWS CLI (warning only - required later for CodeArtifact access, but not needed during installation)
The installer will automatically check these prerequisites before proceeding.
Option 1: Quick Install (Recommended)
Install directly from GitHub with a single command. You can use either command-line arguments (recommended) or interactive prompts.
Important: Run the script from the parent directory where you want to create the service. For example:
To create
/git/mcp-my-service/, run the script from/git/directoryThe script will create
mcp-{service-name}/in the current working directory
Direct Installation with Arguments (Recommended)
The simplest way - directly download and execute with parameters:
# Navigate to the parent directory where you want to create the service
cd /git
# Run the installer
curl -sSL https://raw.githubusercontent.com/OxSci-AI/oxsci-mcp-scaffold/main/install.py | python3 - \
--service-name document-processor \
--tool-name document_processor \
--yesCommand-line options:
--service-name: Service name (must start with a letter, lowercase letters/numbers/hyphens only)--tool-name: Initial tool name (must start with a letter, lowercase letters/numbers/underscores only)--yesor-y: Skip confirmation prompt (recommended for non-interactive mode)--skip-env-check: Skip environment prerequisites check (not recommended)
Interactive Mode
If you prefer to answer prompts interactively, download the script first:
# Navigate to the parent directory where you want to create the service
cd /git
# Download and run interactively
curl -sSL https://raw.githubusercontent.com/OxSci-AI/oxsci-mcp-scaffold/main/install.py > install.py
python3 install.pyThe script will:
Check environment prerequisites (Python 3.11+, Git, network connectivity)
Show current directory and confirm where the service will be created
Prompt for service name (e.g.,
document-processor- will createmcp-document-processordirectory)Prompt for tool name (e.g.,
document_processor- will create the initial tool file and configure it)
Note: If you try to use the pipe method without arguments, you'll get an
EOFErrorbecause stdin is redirected. Always provide--service-nameand--tool-namewhen using the pipe method.
The installer will:
Verify environment prerequisites (Python 3.11+, Git, network)
Download the latest scaffold template from GitHub
Create a new project directory with all files configured
Initialize a git repository
Set up the tool structure and configuration files
Option 2: Manual Setup
If you prefer to clone the repository first:
# Clone the repository
git clone https://github.com/OxSci-AI/oxsci-mcp-scaffold.git
cd oxsci-mcp-scaffold
# Run setup script
python setup.pyThe setup script will prompt you for:
Service name (will create
mcp-{service-name}directory)Tool name (will create the initial tool file and configure it)
Quick Start
After installation, navigate to your new service directory:
cd mcp-{your-service-name}1. Configure AWS CodeArtifact
Before installing dependencies, configure access to the private package repository:
# Make the script executable (if not already)
chmod +x entrypoint-dev.sh
# Run the configuration script (valid for 12 hours)
./entrypoint-dev.shNote: The authentication token expires after 12 hours. Re-run this script when needed.
2. Install Dependencies
poetry install3. Development
Run Server Locally
poetry run uvicorn app.core.main:app --reload --port 8060Access the server at: http://localhost:8060
API documentation: http://localhost:8060/docs
Tool discovery: http://localhost:8060/tools/discover
Tool list: http://localhost:8060/tools/list
Test Your Tools
Check server status:
curl http://localhost:8060/Discover available tools:
curl http://localhost:8060/tools/discoverList all tools (including disabled ones):
curl http://localhost:8060/tools/listExecute a tool:
curl -X POST http://localhost:8060/tools/example_tool \
-H "Content-Type: application/json" \
-d '{
"arguments": {
"input_text": "Hello World",
"uppercase": true
},
"context": {
"user_id": "user123"
}
}'Project Structure
mcp-{service-name}/
├── .github/
│ └── workflows/
│ └── docker-builder.yml # CI/CD workflow for building and deploying
├── app/
│ ├── core/
│ │ ├── __init__.py
│ │ ├── config.py # Service configuration
│ │ └── main.py # FastAPI application entry point
│ └── tools/
│ ├── __init__.py # Import tools here
│ ├── example_tool.py # Example tool implementation
│ └── tool_template.py # Comprehensive tool template (disabled)
├── docs/
│ └── TOOL_DEVELOPMENT.md # Tool development guide
├── tests/
│ └── test_example.py # Example tests
├── .vscode/
│ └── extensions.json # Recommended VS Code extensions
├── Dockerfile # Multi-stage Docker build
├── entrypoint-dev.sh # CodeArtifact configuration script
├── install.py # Installer script (for creating new services)
├── setup.py # Setup script (for local setup)
├── pyproject.toml # Project dependencies and configuration
└── README.md # This fileNext Steps
Once your service is set up and running, refer to these resources:
Tool Development Guide: Learn how to create, configure, and deploy MCP tools
API Documentation: Visit http://localhost:8060/docs when running locally
Example Tool: Check
app/tools/example_tool.pyfor a working exampleTemplate Tool: See
app/tools/tool_template.pyfor comprehensive patterns
Tools and Frameworks
This template integrates:
FastAPI: Web framework for building APIs
oxsci-oma-mcp: MCP protocol implementation and tool router
oxsci-shared-core: Shared utilities and configuration (optional)
Related Projects
oxsci-oma-mcp: MCP protocol package
oxsci-oma-core: OMA agent framework
oxsci-shared-core: Shared utilities
License
© 2025 OxSci.AI. All rights reserved.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Model Context Protocol server for the Apideck Unified API. Connect any MCP-compatible agent framework to 100+ accounting systems, HRIS platforms, file storage providers, and more through one integration. More information https://www.apideck.com/mcp-server
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
A Model Context Protocol server for Wix AI tools