TechWithTy / glama-mcp-sdk

Glama MCP SDK for Python/FastAPI: secure client + typed API. TLS, timeouts, retries, pooling. Access 5,000+ MCP servers or run as an MCP server. Optimized for www.glama.ai

Glama MCP SDK (Third-Party Integration)

Production-ready Glama MCP SDK integration for Python/FastAPI inside the Deal Scale backend. This module provides a secure client, typed API schemas, and utilities to connect to the Glama MCP API.

Optimized for www.cybershoptech.com integrations and modern cloud deployments.

Features

• Secure HTTP client with TLS verification, timeouts, retries, and connection pooling
• Typed request/response helpers in api/
• Self-contained config via environment variables (config.py)
• Easy submodule packaging for reuse across services

MCP Capabilities

• Access to 5,000+ Model Context Protocol (MCP) servers and tools via compatible providers/directories.
• Can be wrapped as an MCP server to expose your own tools/endpoints to MCP‑compatible clients.
• Works either as an embedded MCP client inside your FastAPI app or as a standalone MCP component.

High-level steps to turn this SDK into an MCP server:

1. Define an MCP server entrypoint that implements the MCP handshake and tool invocation lifecycle.
2. Expose Glama-backed operations as MCP tools (map to functions in api/ and your app routes).
3. Package and run with your preferred runtime (e.g., uvicorn for HTTP or a process manager) and register with your MCP client.

Directory Structure

• config.py – Environment-driven configuration (GlamaConfig)
• client.py – Placeholder for higher-level client wrapper (extend as needed)
• api/ – Low-level request/response helpers (_requests.py, _responses.py, routes.py, schemas.py)
• .env.example – Example environment configuration
• _tests/ – Place for unit/integration tests

Installation

Add this package to your project (already vendored in backend/app/core/third_party_integrations/glama/). If used externally, install requirements as needed:

pip install httpx pydantic

Configuration (Environment Variables)

Defined in config.py and demonstrated in .env.example:

• GLAMA_BASE_URL (default: https://glama.ai/api/mcp)
• GLAMA_API_KEY (optional)
• GLAMA_VERIFY_TLS (default: true)
• GLAMA_CONNECT_TIMEOUT (default: 3.05)
• GLAMA_READ_TIMEOUT (default: 30.0)
• GLAMA_POOL_CONNECTIONS (default: 20)
• GLAMA_POOL_MAXSIZE (default: 100)
• GLAMA_MAX_RETRIES (default: 3)

Load your .env with your preferred method (e.g., python-dotenv) or via your process manager (Docker, systemd, Render, etc.).

Quick Start

Below is a simple example showing how you might initialize config and call a request helper. Adjust imports to match your app layout.

from backend.app.core.third_party_integrations.glama.config import GlamaConfig
from backend.app.core.third_party_integrations.glama.api._requests import make_request  # example helper

cfg = GlamaConfig()
# Example usage (update path/payload to real endpoints in your app):
resp = make_request(
    base_url=cfg.BASE_URL,
    api_key=cfg.API_KEY,
    verify_tls=cfg.VERIFY_TLS,
    connect_timeout=cfg.CONNECT_TIMEOUT,
    read_timeout=cfg.READ_TIMEOUT,
    method="GET",
    path="/health"
)
print(resp.status_code, resp.text)

Best Practices

• Keep secrets out of source control; use .env or your secret manager
• Tune timeouts and retries per environment (dev vs prod)
• Add unit tests in _tests/ to validate request/response contracts

SEO Notes for www.cybershoptech.com

• Keywords: glama mcp sdk, mcp servers, model context protocol, python fastapi integration, secure api client, connection pooling, retries, tls, production-ready
• Positioning: Access 5,000+ MCP servers/tools and optionally run this package as an MCP server to expose your own tools.
• This integration helps teams ship secure, observable third‑party API connections for www.cybershoptech.com-grade workloads.

License

This folder inherits the repository license unless otherwise noted.