More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Django Backend Engineering Standards

    1

    Production Django/DRF playbook: factory architecture, scrub→upsert→publish pipeline, Keycloak auth, Celery, safe Postgres migrations.

    Free

    1 useSecurity scanned
    Django Backend Engineering Standards

    Django Backend Engineering Standards

    Example session with this skill installed

    Add an endpoint that lets a partner push product listings in bulk. They may resend the same payload, so it has to be safe to retry. Follow our Django backend standards.

    • Read your context and instructions
    • Compiled the django backend engineering

    Summary

    Added a bulk listings endpoint that is safe to retry. Example output: the exact file layout depends on your service style (factory or layered), which the skill checks first.

    What the agent did

    StepDecisionWhy
    Identify styleLayered service, so view -> service -> repositoryMatch the file being edited, then the repo
    ValidateScrubber runs before any write and returns clean DTOsBad rows are rejected early with clear errors
    PersistOne bulk upsert keyed on (partner_id, external_id)A resend updates rows instead of duplicating them
    PublishA Celery task serializes and indexes the changed idsThe request stays fast, the task is idempotent and retried
    AuthPartner API key, private-by-default permission classNo endpoint is public by accident

    Core of the change

    Listing.objects.bulk_create(
        rows,
        update_conflicts=True,
        unique_fields=["partner_id", "external_id"],
        update_fields=["title", "price", "updated_at"],
    )
    publish_listings.delay(ids=changed_ids)  # idempotent, retried with backoff
    

    Checks before merge

    • No N+1 on the serializer context (prefetch the related rows once)
    • Migration for the unique constraint uses a concurrent index
    • Structured log line with partner id and row counts (structlog)

    Next steps

    1. Add an API test that posts the same payload twice and asserts the row count is unchanged.
    2. Add a Celery test for the retry path.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Stop re-deciding how a Django backend should be built

    This skill gives your coding agent a complete, opinionated rulebook for production Django and DRF services. It doesn't teach Django. It tells the agent which pattern to use, why, and what the bad alternative looks like, so the code it writes matches a mature codebase on the first try.

    How it helps

    • Consistent architecture. The agent works out whether a service is multi-brand factory style or layered/repository style, then follows that style instead of mixing both.
    • Safer data writes. Validate, upsert, publish: PostgreSQL-native ON CONFLICT upserts, then serialize to the index and cache.
    • Fewer incidents. Timeout-aware migrations, concurrent indexes, and idempotent Celery tasks with retries.
    • Less review back-and-forth. Naming, exceptions, structured logging (structlog) and test conventions are spelled out.
    • Auth that fits. Keycloak authentication, a Principal permission model, and API keys for service-to-service calls.

    What's inside

    A short SKILL.md router plus 10 reference guides the agent loads only when it touches that area: architecture, API and views, serializers and scrubbers, models and migrations, queries and database, auth and Keycloak, Celery and jobs, caching and search, logging and observability, and code style and testing.

    Every practice explains why it exists, where it's used, when to use it, the tradeoffs, and a production example.

    Use it when

    Building or reviewing Django/DRF features (views, models, serializers, tasks, queries, auth, caching, search) or scaffolding a new service.

    Requires the project to use PostgreSQL. Assumes presence of custom Celery and migration wrappers. Scope is limited to Django/DRF ecosystems.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Permissions

    Network Access
    Environment Variables
    Write Files
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    This skill enforces production-hardened patterns like multi-brand factory architectures, native PostgreSQL optimizations, and lock-safe migrations that standard AI models often overlook or misconfigure.

    It is optimized for use with advanced coding assistants like GitHub Copilot, Cursor, and Windsurf, specifically when they are configured to follow custom project instructions or .cursorrules files.

    Yes, while the skill promotes high-performance PostgreSQL and Celery patterns, it is designed to be injected into existing codebases to refactor legacy views and optimize current database queries.

    The skill requires a PostgreSQL database environment and assumes the use of Django REST Framework; it also performs best when structured logging and Celery are part of your stack.

    You receive the complete logic set for enforcing these engineering standards, including architectural checklists, PR review guidelines, and scaffold templates for the supported patterns.

    The skill is periodically updated to reflect the latest stable releases of Django and DRF, as well as evolving best practices for high-concurrency PostgreSQL management.

    New in APIs & Backend