🔬 Aignostics Python SDK

Tip

📚 Online documentation - 📖 PDF Manual

Note

🧠 This project was scaffolded using the template oe-python-template with copier.

---

Python SDK providing access to Aignostics AI services.

Scaffolding

This Copier template enables you to quickly generate (scaffold) a Python package with fully functioning build and test automation:

1. Projects generated from this template can be easily updated to benefit from improvements and new features of the template.
2. During project generation, you can flexibly configure naming of the Python distribution, import package, main author, GitHub repository, organization, and many other aspects to match your specific requirements (see copier.yml for all available options).

Development Infrastructure

Projects generated with this template come with a comprehensive development toolchain and quality assurance framework that supports the entire software development lifecycle - from coding and testing to documentation, release management, and compliance auditing. This infrastructure automates routine tasks, enforces code quality standards, and streamlines the path to production:

1. Linting with Ruff
2. Static type checking with mypy
3. Complete set of pre-commit hooks including detect-secrets and pygrep
4. Unit and E2E testing with pytest including parallel test execution
5. Matrix testing in multiple environments with nox
6. Test coverage reported with Codecov and published as release artifact
7. CI/CD pipeline automated with GitHub Actions
8. CI/CD pipeline can be run locally with act
9. Code quality and security checks with SonarQube and GitHub CodeQL
10. Dependency monitoring and vulnerability scanning with pip-audit, trivy, Renovate, and GitHub Dependabot
11. Licenses of dependencies extracted with pip-licenses, matched with allow list, and published as release artifacts in CSV and JSON format for further compliance checks
12. Generation of attributions from extracted licenses
13. Software Bill of Materials (SBOM) generated in CycloneDX and SPDX formats with cyclonedx-python resp. trivy, published as release artifacts
14. Version and release management with bump-my-version
15. Changelog and release notes generated with git-cliff
16. Documentation generated with Sphinx including reference documentation and PDF export
17. Documentation published to Read The Docs
18. Interactive OpenAPI specification with Swagger
19. Python package published to PyPI
20. Docker images published to Docker.io and GitHub Container Registry with artifact attestations
21. One-click development environments with Dev Containers and GitHub Codespaces
22. Settings for use with VSCode
23. Settings and custom instructions for use with GitHub Copilot

Application Features

Beyond development tooling, projects generated with this template include the code, documentation, and configuration of a fully functioning demo application and service. This reference implementation serves as a starting point for your own business logic with modern patterns and practices already in place:

1. Service architecture suitable for use as shared library
2. Validation with pydantic
3. Command-line interface (CLI) with Typer
4. Versioned Web API with FastAPI
5. Interactive Jupyter notebook and reactive Marimo notebook
6. Simple Web UI with Streamlit
7. Configuration to run the CLI and API in a Docker container including setup for Docker Compose
8. Documentation including badges, setup instructions, contribution guide and security policy
9. Preparation to deploy API as serverless function to Vercel

Explore here for what's generated out of the box.

Generate a new project

To generate, build and release a fully functioning project in a few minutes, follow these 5 steps:

Step 1: Execute the following command to install or update tooling.

# Install Homebrew, uv package manager, copier and further dev tools
curl -LsSf https://raw.githubusercontent.com/helmut-hoffer-von-ankershoffen/oe-python-template/HEAD/install.sh | sh

Step 2: Create a repository on GitHub, clone to your local machine, and change into it's directory.

Step 3: Execute the following command to generate a new project based on this template.

# Ensure to stand in your freshly created git repository before executing this command
copier copy --trust gh:helmut-hoffer-von-ankershoffen/oe-python-template .

Step 4: Execute the following commands to push your initial commit to GitHub.

git add .
git commit -m "chore: Initial commit"
git push

Check the Actions tab of your GitHub repository: The CI/CD workflow of your project is already running!

The workflow will fail at the SonarQube step, as this external service is not yet configured for our new repository. We will configure SonarQube and other services in the next step!

Notes:

1. Check out this manual on how to set up signed commits

Step 5: Follow the instructions to wire up external services such as CloudCov, SonarQube Cloud, Read The Docs, Docker.io, and Streamlit Community Cloud.

Step 6: Release the first version of your project

make bump

Notes:

1. You can remove the above sections - from "Scaffolding" to this notes - post having successfully generated your project.
2. The following sections refer to the dummy application and service generated into the tests and src folder by this template. Use the documentation and code as inspiration, adapt to your business logic, or remove and start documenting and coding from scratch.

Overview

Adding Aignostics Python SDK to your project as a dependency is easy. See below for usage examples.

uv add aignostics             # add dependency to your project

If you don't have uv installed follow these instructions. If you still prefer pip over the modern and fast package manager uv, you can install the library like this:

pip install aignostics        # add dependency to your project

Executing the command line interface (CLI) in an isolated Python environment is just as easy:

uvx aignostics hello-world       # prints "Hello, world! [..]"
uvx aignostics serve             # serves web API
uvx aignostics serve --port=4711 # serves web API on port 4711

Notes:

1. The API is versioned, mounted at /api/v1 resp. /api/v2
2. While serving the web API go to http://127.0.0.1:8000/api/v1/hello-world to see the respons of the hello-world operation.
3. Interactive documentation is provided at http://127.0.0.1:8000/api/docs

The CLI provides extensive help:

uvx aignostics --help                # all CLI commands
uvx aignostics hello-world --help    # help for specific command
uvx aignostics echo --help
uvx aignostics openapi --help
uvx aignostics serve --help

Operational Excellence

This project is designed with operational excellence in mind, using modern Python tooling and practices. It includes:

1. Various examples demonstrating usage: a. Simple Python script b. Streamlit web application deployed on Streamlit Community Cloud c. Jupyter and Marimo notebook
2. Complete reference documentation on Read the Docs
3. Transparent test coverage including unit and E2E tests (reported on Codecov)
4. Matrix tested with multiple python versions to ensure compatibility (powered by Nox)
5. Compliant with modern linting and formatting standards (powered by Ruff)
6. Up-to-date dependencies (monitored by Renovate and Dependabot)
7. A-grade code quality in security, maintainability, and reliability with low technical debt and codesmell (verified by SonarQube)
8. Additional code security checks using CodeQL
9. Security Policy
10. License compliant with the Open Source Initiative (OSI)
11. 1-liner for installation and execution of command line interface (CLI) via uv(x) or Docker
12. Setup for developing inside a devcontainer included (supports VSCode and GitHub Codespaces)

Usage Examples

The following examples run from source - clone this repository using git clone git@github.com:aignostics/python-sdk.git.

Minimal Python Script:

"""Example script demonstrating the usage of the service provided by Aignostics Python SDK."""

from dotenv import load_dotenv
from rich.console import Console

from aignostics import Service

console = Console()

load_dotenv()

message = Service.get_hello_world()
console.print(f"[blue]{message}[/blue]")

Show script code - Read the reference documentation

Streamlit App

Serve the functionality provided by Aignostics Python SDK in the web by easily integrating the service into a Streamlit application.

Try it out! - Show the code

... or serve the app locally

uv sync --all-extras                                # Install streamlit dependency part of the examples extra, see pyproject.toml
uv run streamlit run examples/streamlit.py          # Serve on localhost:8501, opens browser

Notebooks

Jupyter

Show the Jupyter code

... or run within VSCode

uv sync --all-extras                                # Install dependencies required for examples such as Juypyter kernel, see pyproject.toml

Install the Jupyter extension for VSCode

Click on examples/notebook.ipynb in VSCode and run it.

Marimo

Show the marimo code

Execute the notebook as a WASM based web app

uv sync --all-extras                                # Install ipykernel dependency part of the examples extra, see pyproject.toml
uv run marimo run examples/notebook.py --watch      # Serve on localhost:2718, opens browser

or edit interactively in your browser

uv sync --all-extras                                # Install ipykernel dependency part of the examples extra, see pyproject.toml
uv run marimo edit examples/notebook.py --watch     # Edit on localhost:2718, opens browser

... or edit interactively within VSCode

Install the Marimo extension for VSCode

Click on examples/notebook.py in VSCode and click on the caret next to the Run icon above the code (looks like a pencil) > "Start in marimo editor" (edit).

Command Line Interface (CLI)

Run with uvx

Show available commands:

uvx aignostics --help

Execute commands:

uvx aignostics hello-world
uvx aignostics echo --help
uvx aignostics echo "Lorem"
uvx aignostics echo "Lorem" --json
uvx aignostics openapi
uvx aignostics openapi --output-format=json
uvx aignostics serve

Environment

The service loads environment variables including support for .env files.

cp .env.example .env              # copy example file
echo "THE_VAR=MY_VALUE" > .env    # overwrite with your values

Now run the usage examples again.

Run with Docker

You can as well run the CLI within Docker.

docker run helmuthva/aignostics-python-sdk --help
docker run helmuthva/aignostics-python-sdk hello-world
docker run helmuthva/aignostics-python-sdk echo --help
docker run helmuthva/aignostics-python-sdk echo "Lorem"
docker run helmuthva/aignostics-python-sdk echo "Lorem" --json
docker run helmuthva/aignostics-python-sdk openapi
docker run helmuthva/aignostics-python-sdk openapi --output-format=json
docker run helmuthva/aignostics-python-sdk serve

Execute command:

docker run --env THE_VAR=MY_VALUE helmuthva/aignostics-python-sdk echo "Lorem Ipsum"

Or use docker compose

The .env is passed through from the host to the Docker container.

docker compose run aignostics --help
docker compose run aignostics hello-world
docker compose run aignostics echo --help
docker compose run aignostics echo "Lorem"
docker compose run aignostics echo "Lorem" --json
docker compose run aignostics openapi
docker compose run aignostics openapi --output-format=json
echo "Running Aignostics Python SDK's API container as a daemon ..."
docker compose up -d
echo "Waiting for the API server to start ..."
sleep 5
echo "Checking health of v1 API ..."
curl http://127.0.0.1:8000/api/v1/healthz
echo ""
echo "Saying hello world with v1 API ..."
curl http://127.0.0.1:8000/api/v1/hello-world
echo ""
echo "Swagger docs of v1 API ..."
curl http://127.0.0.1:8000/api/v1/docs
echo ""
echo "Checking health of v2 API ..."
curl http://127.0.0.1:8000/api/v2/healthz
echo ""
echo "Saying hello world with v1 API ..."
curl http://127.0.0.1:8000/api/v2/hello-world
echo ""
echo "Swagger docs of v2 API ..."
curl http://127.0.0.1:8000/api/v2/docs
echo ""
echo "Shutting down the API container ..."
docker compose down

Extra: Lorem Ipsum

Dolor sit amet, consectetur adipiscing elit. Donec a diam lectus. Sed sit amet ipsum mauris. Maecenas congue ligula ac quam.

Further Reading

• Inspect our security policy with detailed documentation of checks, tools and principles.
• Our release notes provide a complete log of recent improvements and changes.
• Check out the reference with detailed documentation of public classes and functions.
• In case you want to help us improve 🔬 Aignostics Python SDK: The contribution guidelines explain how to setup your development environment and create pull requests.
• We gratefully acknowledge the open source projects that this project builds upon. Thank you to all these wonderful contributors!

Star History