# Mayrlabs Framework Django > Django rules focused on class-based views, clean ORM usage, and decoupling fat models where appropriate. - Skill: `tomevault-io/mayrlabs-framework-django` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/mayrlabs-framework-django` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/mayrlabs-framework-django/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Coding & Dev Tools - License: MIT - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/mayrlabs-framework-django --- # MayR Labs Django Doctrine ## Purpose To maintain robust, predictable Django architectures by enforcing clear patterns over raw implicit magic, and preventing the "fat models, fat views" anti-pattern. ## Audience AI Agents generating Django views, serializers, and models. ## Core Rules & Constraints ### 1. Views & Routing - Leverage Class-Based Views (CBVs) or ViewSets (in Django Rest Framework). Avoid sprawling function-based views containing complex logic. - Do not place heavy multi-step business logic inside a View. Abstract it into a `services.py` or `selectors.py` pattern (akin to separating queries from mutations). ### 2. ORM Optimization - **N+1 Prevention**: Radically enforce the use of `select_related()` (for foreign keys) and `prefetch_related()` (for many-to-many/reverse relations) on QuerySets that will be serialized or iterated over. ### 3. Security - Never expose internal database fields via `fields = '__all__'` on generic Serializers or Forms. Fields must be explicitly listed. ## Inputs & Outputs - **Input**: Django `views.py`, `models.py`, `serializers.py`. - **Output**: Service-layered, optimized, and explicit Django architectures. --- > Source: [MayR-Labs/skills](https://github.com/MayR-Labs/skills) — distributed by [TomeVault](https://tomevault.io).