Social Network Kata¶

A small, Docker-packaged social-network backend built test-first with strict TDD discipline. The kata spans three bounded contexts — Users, Posts / Timeline, and Messaging — and is intentionally minimal: in-memory repositories, no web framework, no real database. The value is in the process (requirements → architecture → RED-GREEN-REFACTOR → CI/CD → docs), not in the line count.

This repository is the coursework for the SDP — Powered by AI Agents course at the University of Novi Sad.

What the kata solves¶

Every user-facing feature is captured as a bundle (user story + supporting INFRA / BE / FE stories) with GIVEN-WHEN-THEN scenarios. Each scenario becomes one test; each test drives one minimal implementation commit. The git log is the primary artefact — it shows the incremental progression from empty repo to working feature.

Tech stack¶

Concern	Choice
Language / runtime	Python 3.12
Testing	pytest
Packaging / execution	Docker (`python:3.12-slim`)
CI	GitHub Actions (tests on every push)
Docs	Sphinx + MyST parser, deployed to GitHub Pages
Architecture method	arc42 + C4 diagrams (PlantUML)
Pre-commit hooks	black, isort, ruff, bandit, detect-secrets, docker-test

Architecture overview¶

Three bounded contexts communicate over clear interfaces and own their own data:

Users Service — registration, login, profile, follow graph
Posts Service — create post, aggregated timeline feed
Messaging Service — direct messages between users

The deployment topology is Docker containers behind an nginx reverse proxy. Full detail lives in the arc42 documentation — see chapter 5 (Building Block View) and chapter 7 (Deployment View) first.

Build and run¶

All commands assume Docker is available on the host.

# Build the image
docker build -t social-network-kata .

# Run the scripted end-to-end demo (default)
docker run --rm social-network-kata

# Run the test suite inside the container
docker run --rm social-network-kata pytest tests/ -v

The default container command is python -m src.demo — a scripted walkthrough of the three bounded contexts that prints each step with its originating Story/Scenario ID. The test suite is available by overriding the command, which is what CI runs on every push.

Run tests locally (without Docker)¶

pip install -r requirements.txt
pytest tests/ -v

Project structure¶

.
├── Dockerfile                  # Runtime image; CMD runs the test suite
├── requirements.txt            # Runtime + test dependencies
├── src/
│   ├── users/                  # Users bounded context
│   ├── posts/                  # Posts / Timeline bounded context
│   └── messaging/              # Messaging bounded context
├── tests/
│   ├── users/                  # User registration, login, profile scenarios
│   ├── posts/                  # Post creation and timeline scenarios
│   └── messaging/              # Send message, read conversation, @mention scenarios
├── docs/
│   ├── architecture/           # arc42 chapters 1–12
│   ├── user-stories/           # Story bundles per bounded context
│   ├── conf.py                 # Sphinx configuration
│   └── index.rst               # Documentation entry point
├── .github/workflows/          # CI (tests) and docs-deploy pipelines
├── .kiro/agents/               # Agent configurations used during the course
└── scripts/hooks/              # Pre-commit hook implementations

Documentation¶

The Sphinx site is published to GitHub Pages and covers the full arc42 architecture plus every user-story bundle with GIVEN-WHEN-THEN scenarios.

Live site: https://ivanovic0245.github.io/sdp-powered-by-ai-agents-aleksandar-ivanovic/

To build the docs locally:

pip install -r docs/requirements.txt
cd docs && make html
# Open docs/_build/html/index.html

License¶

Released under the MIT License .

Author¶

Aleksandar Ivanović — student, SDP — Powered by AI Agents, University of Novi Sad, 2026.