Atom Api

Record CRUD API (Atom API - First Commit)

Goal

Build a minimal, production-structured CRUD API using FastAPI that supports a single Record model with basic create and read operations. The system should establish clean architecture foundations (controller → service → repository), in-memory storage, and simple validation, preparing the codebase for future persistence and feature expansion.

Changes Needed

• Add FastAPI application bootstrap (app/main.py)
• Add project folder structure:
  • app/controllers/
  • app/services/
  • app/repositories/
  • app/models/
  • app/schemas/
  • app/core/ (config, settings)
• Add environment configuration using .env + Pydantic settings
• Define Record data model:
  • id (string/UUID)
  • model (string)
  • title (string)

  • json (dict

    optional)

  • markdown (string

    optional)

  • created (datetime)
  • updated (datetime)
• Implement in-memory storage (dictionary or list in repository layer)
• Add repository layer:
  • create_record
  • get_record_by_id
  • list_records
• Add service layer:
  • validation rules (required: model, title)
  • timestamp handling (created/updated)
  • business logic delegation to repository
• Add controller layer (FastAPI routers):
  • POST /model
  • GET /model/{id}
  • GET /model
• Add Pydantic schemas:
  • RecordCreate
  • RecordResponse
• Add basic error handling:
  • 404 for missing record
  • 400 for invalid input
• Add FastAPI app entrypoint with router registration
• Add minimal README:
  • setup instructions
  • env variables
  • how to run server
  • sample curl requests

Acceptance Criteria

• FastAPI server starts successfully with uvicorn
• Project follows layered architecture (controller → service → repository)
• Record model is defined with required fields and timestamps
• POST /model creates a record and returns it with generated id
• GET /model/{id} returns a single record or 404 if not found
• GET /model returns list of all stored records
• Data is stored in-memory via repository abstraction
• Required field validation (model, title) is enforced
• Basic error handling returns correct HTTP status codes (400, 404)
• Environment configuration is loaded via .env
• README includes clear setup and usage instructions