Monorepo Management (Nx/Turborepo) Cheat Sheet

Covers workspace structure, task pipelines, remote caching, affected-project detection, and dependency graphs for Nx and Turborepo.

2 PagesIntermediateApr 1, 2026

Turborepo `turbo.json` Pipeline

Define task dependencies and cache behavior once; Turborepo topologically orders and caches runs.

json

{  "$schema": "https://turbo.build/schema.json",  "tasks": {    "build": {      "dependsOn": ["^build"],      "outputs": ["dist/**", ".next/**"]    },    "test": {      "dependsOn": ["build"],      "outputs": []    },    "lint": {      "outputs": []    },    "dev": {      "cache": false,      "persistent": true    }  }}

Turborepo CLI Commands

Run tasks across the whole workspace or scoped to changed packages.

bash

# run build in every package that has a build script, respecting dependsOn orderturbo run build# only packages affected since the given git refturbo run test --filter=...[origin/main]# run for one package and its dependentsturbo run build --filter=@acme/ui...# remote caching (Vercel by default, or self-hosted)turbo loginturbo linkturbo run build --remote-only

Nx `nx.json` & Project Graph

Nx infers a project graph from imports and configures cacheable target defaults centrally.

json

{  "targetDefaults": {    "build": {      "dependsOn": ["^build"],      "cache": true,      "inputs": ["production", "^production"]    },    "test": {      "cache": true,      "inputs": ["default", "^production"]    }  },  "namedInputs": {    "default": ["{projectRoot}/**/*"],    "production": ["!{projectRoot}/**/*.spec.ts"]  }}

Nx CLI Commands

Nx's `affected` commands use the project graph to run only what changed.

bash

# visualize the dependency graph in the browsernx graph# run a target for one projectnx build my-app# run a target for every project affected by uncommitted/branch changesnx affected -t build test lint --base=main# run with Nx Cloud distributed task execution + remote cachingnx affected -t build --parallel=5

Nx vs Turborepo Quick Comparison

Both solve caching/orchestration; they differ in scope and opinionation.

Turborepo- lightweight task runner/cache, config in turbo.json, framework-agnostic, minimal opinions
Nx- full toolkit: generators, plugins, project graph visualization, code-owned migrations
Remote caching- both support it (Vercel Remote Cache / Nx Cloud or self-hosted alternatives)
Affected detection- turbo uses --filter=...[ref]; Nx uses `nx affected` with --base/--head
Task pipeline config- turbo.json `tasks` vs nx.json `targetDefaults`, both express dependsOn graphs
Best fit- Turborepo for simpler JS/TS monorepos; Nx for large multi-framework orgs wanting generators/enforced boundaries

Pro Tip

Always scope CI to affected/changed projects (`turbo run test --filter=...[origin/main]` or `nx affected -t test --base=origin/main`) instead of running every task on every push — this is the single biggest lever for keeping monorepo CI times flat as the repo grows.

Was this cheat sheet helpful?

CQRS & Event Sourcing Cheat Sheet

Optuna Hyperparameter Optimization Cheat Sheet

Explore Topics

#MonorepoManagementNxTurborepo#MonorepoManagementNxTurborepoCheatSheet#Programming#Intermediate#Turborepo#Turbo#Json#Pipeline#DataStructures#DevOps#CheatSheet#SkillVeris

Sri Hayavadhana Info-Tech

Professional Web Designing Services