deps

Command /architect:deps

Analyze module dependencies at three levels: modules, packages, and layers.

Instructions

Step 1: Discover Modules

Find all modules in the project:

Glob: **/modules/*/__init__.py
Glob: **/apps/*/__init__.py

List module structure:

ls -la {{ path | default: "src/modules" }}

Step 2: Analyze Module Dependencies

For each module, extract import statements:

# Pattern to find cross-module imports
Grep: "from src.modules." in each module
Grep: "import src.modules." in each module

Build dependency matrix:

Module	Depends On	Dependents
users	core	orders, payments
orders	users, products	payments
payments	orders	-

Step 3: Detect Circular Dependencies

A circular dependency exists when:

• Module A imports Module B
• Module B imports Module A (directly or transitively)

Detection algorithm:

1. Build directed graph of module dependencies
2. Find strongly connected components (SCC)
3. SCC with >1 node = circular dependency

Common circular patterns:

# Direct circular
# users/services.py
from src.modules.orders.services import OrderService

# orders/services.py
from src.modules.users.services import UserService  # Circular!

# Transitive circular
# A → B → C → A

Step 4: Analyze Package Dependencies

Check external package dependencies:

# From pyproject.toml or requirements.txt
cat pyproject.toml | grep -A 100 "\[project.dependencies\]"

Categorize dependencies:

• Core: Python stdlib, typing
• Framework: FastAPI, Pydantic, SQLAlchemy
• Infrastructure: asyncpg, redis, celery
• Development: pytest, ruff, mypy

Step 5: Analyze Layer Dependencies

Verify layer dependency rules:

Presentation → Application → Domain ← Infrastructure

Violations to detect:

• Router importing Repository directly (should go through Service)
• Service importing Router (reverse dependency)
• Model importing Service (domain shouldn't know about application)