domains.json

domains.json captures the DDD bounded context map of the legacy system. It tells agents which parts of the codebase belong together, how tightly coupled they are, and what their upstream/downstream relationships look like.

Full Example

{
  "contexts": [
    {
      "name": "Accounts",
      "type": "core",
      "entities": ["GLEntry", "SalesInvoice", "PurchaseInvoice", "PaymentEntry", "JournalEntry"],
      "aggregateRoot": "GLEntry",
      "loc": 45000,
      "endpoints": 89,
      "coupling": {
        "internal": 0.82,
        "external": 0.34
      },
      "dependencies": ["Stock", "Selling", "Buying"],
      "godClasses": [
        { "file": "controllers/accounts_controller.py", "loc": 4412, "functions": 168 }
      ]
    },
    {
      "name": "Stock",
      "type": "supporting",
      "entities": ["StockEntry", "StockLedgerEntry", "Warehouse", "Item"],
      "aggregateRoot": "StockEntry",
      "loc": 28000,
      "endpoints": 62,
      "coupling": {
        "internal": 0.75,
        "external": 0.41
      },
      "dependencies": ["Accounts"],
      "godClasses": [
        { "file": "stock/doctype/stock_entry/stock_entry.py", "loc": 4149, "functions": 120 }
      ]
    },
    {
      "name": "Setup",
      "type": "generic",
      "entities": ["Company", "Currency", "Country"],
      "aggregateRoot": "Company",
      "loc": 8000,
      "endpoints": 15,
      "coupling": {
        "internal": 0.90,
        "external": 0.12
      },
      "dependencies": [],
      "godClasses": []
    }
  ],
  "contextMap": [
    { "upstream": "Accounts", "downstream": "Stock", "type": "conformist" },
    { "upstream": "Accounts", "downstream": "Selling", "type": "conformist" },
    { "upstream": "Setup", "downstream": "Accounts", "type": "shared-kernel" }
  ]
}

Interactive Visualization

Field Reference

contexts Array

Each entry describes a bounded context:

Field	Type	Required	Description
name	string	Yes	Context name (e.g., “Accounts”, “Stock”)
type	string	Yes	DDD classification: "core", "supporting", or "generic"
entities	string[]	Yes	Business entities belonging to this context
aggregateRoot	string	Yes	The primary entity (aggregate root) of this context
loc	number	Yes	Lines of code in this context
endpoints	number	Yes	API endpoints owned by this context
coupling	object	Yes	Coupling metrics (see below)
dependencies	string[]	Yes	Other contexts this context depends on
godClasses	array	No	Files exceeding complexity thresholds

coupling Object

Field	Type	Description
internal	number	Internal cohesion score (0.0 to 1.0). Higher is better.
external	number	External coupling score (0.0 to 1.0). Lower is better.

A well-bounded context has high internal cohesion (above 0.7) and low external coupling (below 0.3).

godClasses Array

Field	Type	Description
file	string	Path to the file relative to the project root
loc	number	Lines of code
functions	number	Number of functions/methods

Files with more than 2,000 LOC or 100 functions should be flagged as God-classes. These typically need decomposition before extraction.

contextMap Array

Describes relationships between bounded contexts:

Field	Type	Description
upstream	string	The context that provides data/services
downstream	string	The context that consumes data/services
type	string	Relationship type: "conformist", "acl", "partnership", "shared-kernel"

Context Map Relationship Types

Type	Meaning
conformist	Downstream conforms to upstream’s model without translation
acl	Anti-Corruption Layer — downstream translates upstream’s model into its own
partnership	Both contexts evolve together with shared ownership
shared-kernel	Shared code/model that both contexts depend on

Context Type Guidelines

Type	Description	Extraction Priority
core	Highest business value. Contains the domain logic that differentiates the product.	Extract first (highest value)
supporting	Supports the core domain. Necessary but not differentiating.	Extract second
generic	Generic functionality (auth, config, notifications). Could be replaced by off-the-shelf.	Extract last or replace

Next Steps

• complexity.json — Extraction difficulty tiers
• Context Map Example — Interactive visualization