Harness-engineering nestjs-dependency-injection

NestJS Dependency Injection

install
source · Clone the upstream repo
git clone https://github.com/Intense-Visions/harness-engineering
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/Intense-Visions/harness-engineering "$T" && mkdir -p ~/.claude/skills && cp -r "$T/agents/skills/claude-code/nestjs-dependency-injection" ~/.claude/skills/intense-visions-harness-engineering-nestjs-dependency-injection-232ed4 && rm -rf "$T"
manifest: agents/skills/claude-code/nestjs-dependency-injection/SKILL.md
source content

NestJS Dependency Injection

Master NestJS DI container with tokens, useClass/useValue/useFactory providers

When to Use

  • You need to provide an interface implementation that is swappable (e.g., a real database vs. an in-memory mock)
  • You need to inject a plain value (config object, constant, SDK client instance) that is not a class
  • You need to run async initialization before a provider is usable (factory providers with async/await)
  • You want to understand why "Cannot resolve dependencies" errors occur

Instructions

  1. Standard provider (shorthand):

    providers: [MyService]
    — NestJS creates one singleton instance, injected by class type.

  2. useClass — swap implementations:

providers: [
  { provide: MyService, useClass: process.env.NODE_ENV === 'test' ? MockMyService : MyService },
];
  1. useValue — inject constants or SDK clients:
export const STRIPE_CLIENT = 'STRIPE_CLIENT';

providers: [
  { provide: STRIPE_CLIENT, useValue: new Stripe(process.env.STRIPE_KEY!) }
]

// inject with @Inject token
constructor(@Inject(STRIPE_CLIENT) private stripe: Stripe) {}
  1. useFactory — async initialization:
providers: [
  {
    provide: DATABASE_CONNECTION,
    useFactory: async (config: ConfigService): Promise<DataSource> => {
      const ds = new DataSource({ url: config.get('DATABASE_URL') });
      await ds.initialize();
      return ds;
    },
    inject: [ConfigService],
  },
];
  1. InjectionToken vs string tokens: Prefer
    Symbol
    -based or class-based tokens over plain strings to avoid collisions:
export const MAIL_OPTIONS = new InjectionToken<MailOptions>('MailOptions');
  1. Inject in constructor with
    @Inject(TOKEN)
    when the token is not a class:
constructor(@Inject(MAIL_OPTIONS) private options: MailOptions) {}
  1. Optional injection: Use
    @Optional()
    when a provider may not be registered:
    constructor(@Optional() @Inject(CACHE) private cache?: Cache) {}
    .

Details

NestJS uses a hierarchical IoC container built on top of Reflect metadata. When you add a class to

providers
, the container reads its constructor parameter types via TypeScript's
emitDecoratorMetadata
and resolves each dependency recursively.

Token resolution: A provider token can be a class, a string, a Symbol, or an

InjectionToken<T>
. The container matches the
provide
key to the
@Inject()
token (or the constructor type for class providers). Mismatched tokens are the second most common DI error after missing exports.

Scopes and singleton behavior: Default (singleton) scope means

useFactory
runs once at app startup.
REQUEST
scope runs the factory (or constructor) per request — useful for per-request database connections or tenant-aware clients.

Testing:

overrideProvider(MyService).useValue(mockService)
in
Test.createTestingModule()
replaces any token with a mock without touching the module graph. This is the cleanest way to unit-test controllers and services.

ModuleRef.resolve()
for dynamic resolution: When you need to resolve a provider at runtime (e.g., strategy pattern where the concrete implementation depends on runtime data), inject
ModuleRef
and call
moduleRef.resolve(SomeService)
.

Source

https://docs.nestjs.com/fundamentals/custom-providers

Process

  1. Read the instructions and examples in this document.
  2. Apply the patterns to your implementation, adapting to your specific context.
  3. Verify your implementation against the details and edge cases listed above.

Harness Integration

  • Type: knowledge — this skill is a reference document, not a procedural workflow.
  • No tools or state — consumed as context by other skills and agents.

Success Criteria

  • The patterns described in this document are applied correctly in the implementation.
  • Edge cases and anti-patterns listed in this document are avoided.