1. Conduid
  2. Cloud
  3. Python Fastmcp Azure Appservice
MCP server · Cloud

Python Fastmcp Azure Appservice

Unclaimed cloud
30Low

Scored 6 months ago · breakdown

About Python Fastmcp Azure Appservice

Python Fastmcp Azure Appservice is an MCP server in the Cloud category.

Install

From source
git clone https://github.com/lTarsoly/python-fastmcp-azure-appservice

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about Python Fastmcp Azure Appservice

Powered by Claude · Grounded in docs

I know everything about Python Fastmcp Azure Appservice. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

  • ✓README presentThe repository ships documentation.

5 checks have not been run yet.

README

python-fastmcp-azure-appservice

A minimal example FastMCP server built with FastAPI-style tooling and prepared for deployment to Azure App Service.

This repository contains a tiny FastMCP-based Python app (app.py), a requirements.txt listing runtime dependencies, and a deploy.ps1 script that provisions an Azure App Service and deploys the app.

Contents

  • app.py — example FastMCP application exposing a simple add tool and a root health check route.
  • requirements.txt — Python dependencies needed to run the app.
  • deploy.ps1 — PowerShell script that creates an Azure resource group, App Service plan and Web App, configures startup, packages the app, and deploys it.

Quick contract

  • Inputs: HTTP requests to the web server (tool calls or HTTP GET to /).
  • Outputs: JSON responses; e.g. health check returns { "status": "server running" } and tool calls return tool-specific results.
  • Error modes: missing dependencies or misconfigured environment will prevent the process from starting; Azure deployment errors will be surfaced by the Azure CLI.

Requirements

  • Python 3.11+ (the Azure runtime in deploy.ps1 is set to use Python 3.13; locally, 3.11+ is recommended).
  • pip and virtual environment support.
  • (For deployment) Azure CLI installed and authenticated.

Setup (local)

  1. Create and activate a virtual environment in the repository root:

    PowerShell:

    python -m venv .venv
    .\.venv\Scripts\Activate.ps1
    

    (On macOS / Linux use python -m venv .venv then source .venv/bin/activate.)

  2. Install dependencies:

    python -m pip install --upgrade pip
    python -m pip install -r requirements.txt
    
  3. Run the app locally:

    python app.py
    

    The server listens on 0.0.0.0:8000 by default. Visit http://localhost:8000/ to see the health check JSON.

Example usage

  • Health check (browser or curl):

    curl http://localhost:8000/
    # -> { "status": "server running" }
    
  • Tool example (programmatic): the repository exposes a tool named add(a: int, b: int) -> int which returns the sum of two integers. How you invoke FastMCP tools depends on your client; consult the FastMCP client docs for calling server-side tools.

Deploy to Azure App Service (PowerShell)

The included deploy.ps1 script automates creation of resources and deployment. It assumes you have the Azure CLI installed and are logged in.

  1. Ensure Azure CLI is installed and you are logged in:

    az login
    
  2. Prepare a virtual environment and install dependencies (see Setup above). The deployment script packages the .venv folder, so creating .venv prior to deployment is required by the script as currently written.

  3. Run the deployment script from the repository root in PowerShell:

    .\deploy.ps1
    

The script will:

  • create an Azure resource group, App Service plan, and Web App (runtime set to PYTHON:3.13),
  • configure the web app to run python app.py as its startup command,
  • zip your .venv, app.py, and requirements.txt and deploy them to the web app.

Notes:

  • Adjust resource names and location in deploy.ps1 as needed for your subscription and naming policies.
  • Packaging .venv is convenient but not mandatory; you can instead rely on App Service to build from requirements.txt by changing the script and enabling build-on-deploy.

Troubleshooting

  • If the app fails to start locally, check that dependencies from requirements.txt installed successfully and that you used the correct Python version.

  • For deployment, inspect the App Service logs in the Azure portal or run the following to stream logs:

    az webapp log tail --resource-group LaszlosPythonWebAppResourceGroup --name LaszlosPythonWebApp
    
  • If the web app reports missing packages, consider switching the deployment approach to let App Service build from requirements.txt rather than deploying .venv.

Contributing

Small, focused contributions are welcome. Open an issue or a PR with a clear description of the change.

License

This repository has no license file by default. Add a LICENSE file if you plan to open-source the code.

Contact

Project owner: Laszlo (repository owner lTarsoly)


If you'd like, I can also:

  • add a minimal Makefile or tasks.json for common tasks (run, test, deploy),
  • modify deploy.ps1 to not depend on .venv and instead use App Service build-from-requirements,
  • add a small test that starts app.py and checks the health endpoint programmatically.

README copy captured when this server was indexed; it may lag the repository. The original is authoritative.

Questions

About Python Fastmcp Azure Appservice

How do I install Python Fastmcp Azure Appservice?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is Python Fastmcp Azure Appservice safe to use with an AI agent?

Its trust score is 30 out of 100 (low). It passes 1 of the 1 static security checks run so far. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Python Fastmcp Azure Appservice still maintained?

Conduid hasn't recorded a commit date for this repository yet. Check the repository directly for recent activity.