dependency-alignment

Dependency Version Alignment Skill

This skill resolves dependency versions during parallel decomposition to ensure compatibility with existing project dependencies and avoid transitive conflicts.

When This Skill Activates

• During parallel-decompose Phase 1 (after dependency extraction, before manifest generation)
• When the cto-architect agent identifies dependencies from Tech Spec imports
• When validating manifest.json dependencies before execution

Supported Ecosystems

Ecosystem	Dependency Files	Package Manager	Resolution Tool
Python	pyproject.toml, requirements.txt, uv.lock	uv, pip	uv pip compile
Node.js	package.json, package-lock.json, yarn.lock	npm, yarn, pnpm	npm view, npm ls

Detection Strategy

Python Projects

1. Check for pyproject.toml (preferred - modern Python)
2. Fall back to requirements.txt (legacy)
3. Check for lock files: uv.lock, poetry.lock, requirements.lock

# Detection order
if [[ -f "pyproject.toml" ]]; then
  # Use pyproject.toml as source of truth
elif [[ -f "requirements.txt" ]]; then
  # Fall back to requirements.txt
fi

Node.js Projects

1. Check for package.json (required)
2. Check for lock files: package-lock.json, yarn.lock, pnpm-lock.yaml

# Detection
if [[ -f "package.json" ]]; then
  # Node.js project detected
fi

Resolution Commands

Python with uv

# Step 1: Get currently installed/locked versions
uv pip freeze --format json

# Step 2: Resolve a new dependency against existing project
# Creates a temporary requirements and resolves compatible version
uv pip compile pyproject.toml \
  --extra-requirement "pydantic>=2.0" \
  --dry-run \
  --quiet 2>&1

# Step 3: Get the resolved version for a specific package
uv pip compile - <<< "pydantic>=2.0" --dry-run 2>&1 | grep "pydantic=="

# Alternative: Check latest compatible version
uv pip compile - <<< "pydantic>=2.0,<3.0" --dry-run 2>&1