BlogCraftAI

BlogCraftAI is a teaching-ready reference implementation that mirrors the 30-hour training program outlined in the curriculum. It provides a production-grade Django stack with AI integrations, responsive templates, automation samples, and exhaustive tests so that learners can explore every branch with confidence.

Table of Contents

• Feature Overview
• Architecture
• Getting Started
• Use Case Outcomes
• Key API Endpoints
• AI Prompt Engineering Toolkit
• Automation Blueprints
• Testing Strategy

Feature Overview

Branch	Highlights
Main	Onboarding template, resource checklist, communication guidelines.
Django	Modular project (accounts, blog, ai), PostgreSQL-ready config, JWT auth, audited logging.
Prompt Engineering	AiSuggestionService with structured prompts, timeout handling, schema validation, unit tests.
Frontend	Responsive Django templates, accessibility support, SEO insights panel, pagination, dashboards.
BlogCraftAI	Unified architecture diagram, environment configuration, observability hooks, deployment-ready static handling.
AI Agents	n8n and Make.com automation JSON blueprints for editorial workflows.

Architecture

• Backend: Django 5 + Django REST Framework powering authentication, blog CRUD, and AI endpoints. Custom user model enforces email logins and profile metadata.
• Database: Ships with SQLite for local development and environment-driven PostgreSQL configuration for staging/production.
• Frontend: Server-rendered templates styled with a responsive utility sheet. Components are WCAG-conscious, keyboard navigable, and optimised for Lighthouse/PageSpeed budgets.
• AI Layer: AiSuggestionService encapsulates prompt engineering contracts, input validation, resilience strategies, and JSON parsing into SeoSuggestion dataclasses.
• Automation: Reusable workflow definitions demonstrate how to orchestrate BlogCraftAI APIs from n8n and Make.com for scheduling, enrichment, and notifications.

Getting Started

1. Install dependencies

   pip install -r requirements.txt

2. Apply migrations

   cd backend
   python manage.py migrate

3. Create a superuser (optional but recommended)

   python manage.py createsuperuser

4. Run the development server

   python manage.py runserver

5. Access the app at http://127.0.0.1:8000/

Environment variables such as POSTGRES_DB, AI_PROVIDER_URL, and JWT lifetimes can be configured via the shell or a .env file to align with deployment targets.

Use Case Outcomes

• ✅ Learners can register, authenticate via JWT, and manage their profile through the /accounts endpoints and UI flows.
• ✅ Editors create, publish, edit, and archive posts with tag management, reaction tracking, and permissioned API access.
• ✅ Visitors browse, search, and filter posts, while responsive templates guarantee accessible navigation and descriptive SEO metadata.
• ✅ AI-assisted SEO suggestions surface on post detail pages and through the /api/seo-suggestions/ endpoint with graceful degradation on provider errors.
• ✅ Automation teams can plug BlogCraftAI into scheduling systems using the provided n8n and Make.com workflow blueprints.

Key API Endpoints

Endpoint	Method	Description
/accounts/register/	POST	Create a new user account.
/accounts/token/	POST	Obtain JWT access and refresh tokens.
/accounts/profile/	GET/PATCH	Retrieve or update the authenticated user profile.
/api/posts/	GET/POST	List or create blog posts (supports filtering, search, ordering).
/api/posts/{slug}/publish/	POST	Publish a draft post (author/staff only).
/api/comments/	POST	Add a public comment to a post.
/api/reactions/	POST	Upsert user reactions (like/dislike) on posts.
/api/seo-suggestions/	POST	Generate AI SEO suggestions for content drafts.

Authenticated requests use Bearer tokens issued by SimpleJWT. Throttling, pagination, and filtering follow DRF conventions to stay predictable for learners.

AI Prompt Engineering Toolkit

• backend/ai/services.py encodes the guardrails, schema validation, retry-safe HTTP access, and structured responses required to integrate large language models responsibly.
• backend/ai/prompts/seo_optimization.md documents the canonical prompt, guardrails, and testing checklist used in class to critique prompt design decisions.
• Comprehensive tests in backend/tests/test_ai_service.py simulate success, timeout, and malformed payload scenarios to encourage defensive programming.

Automation Blueprints

• automations/n8n_content_publish.json schedules daily reviews of draft posts, publishes approved entries, and sends Slack notifications.
• automations/make_scenario_blueprint.json shows how to receive Airtable briefs, seed drafts, request AI suggestions, and push a Teams update—all without writing custom code.

Testing Strategy

Run the full suite with:

pytest

The suite covers API contracts, permission boundaries, AI error handling, and HTML rendering smoke tests. Test fixtures rely on factory_boy for expressive setup and include negative as well as failure simulations (timeouts, validation errors).