AGENT SKILL

reflect-metadata-injection

暂无简介

License 许可
Apache-2.0
最近更新
2026-11-13
Stars
★ 13
Forks
2

03 SKILL.md 正文

/reflect-metadata-injection

When to Use

  • Build lightweight dependency injection containers for agent service registries
  • Attach metadata to classes at decoration time for runtime introspection
  • Auto-wire agent dependencies without explicit factory functions
  • Implement decorators that record audit trails or capability declarations

Do NOT use for

  • Simple config injection (use environment variables or [[cli-config-persistence]])
  • Environments where experimentalDecorators is not enabled

tsconfig requirements

{
  "compilerOptions": {
    "experimentalDecorators": true,
    "emitDecoratorMetadata":  true
  }
}

Core metadata API

import 'reflect-metadata'

// Set metadata on a class
Reflect.defineMetadata('role', 'executor', MyClass)

// Get metadata
const role = Reflect.getMetadata('role', MyClass)  // → 'executor'

// Check existence
Reflect.hasMetadata('role', MyClass)  // → true

// List all keys on a target
Reflect.getMetadataKeys(MyClass)  // → ['role', 'design:type', ...]

// Delete
Reflect.deleteMetadata('role', MyClass)

Injectable decorator + DI container

import 'reflect-metadata'

const INJECTABLE_KEY = Symbol('injectable')
const registry = new Map<Function, unknown>()

// Decorator: mark class as injectable
function Injectable(): ClassDecorator {
  return (target) => {
    Reflect.defineMetadata(INJECTABLE_KEY, true, target)
  }
}

// Decorator: declare dependency token
function Inject(token: symbol): ParameterDecorator {
  return (target, _key, index) => {
    const tokens = Reflect.getMetadata('inject:tokens', target) ?? []
    tokens[index] = token
    Reflect.defineMetadata('inject:tokens', tokens, target)
  }
}

// Container: resolve constructor + inject dependencies
function resolve<T>(cls: new (...args: any[]) => T): T {
  if (!Reflect.getMetadata(INJECTABLE_KEY, cls)) {
    throw new Error(`[di] ${cls.name} is not @Injectable`)
  }

  const tokens: symbol[] = Reflect.getMetadata('inject:tokens', cls) ?? []
  const args = tokens.map(tok => registry.get(tok))
  return new cls(...args)
}

// Usage:
const DB_TOKEN = Symbol('db')
registry.set(DB_TOKEN, { query: async (sql: string) => [] })

@Injectable()
class AgentService {
  constructor(@Inject(DB_TOKEN) private db: any) {}
  async listTasks() { return this.db.query('SELECT * FROM tasks') }
}

const svc = resolve(AgentService)

Read design:paramtypes (emitDecoratorMetadata)

@Injectable()
class OrderService {
  constructor(
    private db:     DatabaseService,
    private logger: LoggerService,
  ) {}
}

// After @Injectable() fires, TypeScript emits:
const types = Reflect.getMetadata('design:paramtypes', OrderService)
// → [DatabaseService, LoggerService]
// Allows container to auto-resolve by constructor type

Audit capability decorator

function Capability(name: string): MethodDecorator {
  return (target, key) => {
    const caps: string[] = Reflect.getMetadata('capabilities', target) ?? []
    caps.push(name)
    Reflect.defineMetadata('capabilities', caps, target)
  }
}

class SandboxAgent {
  @Capability('file-read')
  readFile(path: string) { /* ... */ }

  @Capability('network-fetch')
  fetch(url: string)     { /* ... */ }
}

const caps = Reflect.getMetadata('capabilities', SandboxAgent.prototype)
// → ['file-read', 'network-fetch']
// yamtam can gate these against the agent's declared capability set

Anti-Fake-Pass Checklist

❌ Missing `import 'reflect-metadata'` at entry point → Reflect.defineMetadata is undefined
❌ emitDecoratorMetadata: false → design:paramtypes always undefined, DI auto-wiring fails
❌ Metadata defined on instance vs prototype → Reflect.getMetadata(key, instance) ≠ Reflect.getMetadata(key, Class.prototype)
❌ Symbol tokens not shared between Inject() and registry → tokens don't match, injection returns undefined
❌ resolve() before registry.set() → dependency is undefined, constructor receives undefined silently
❌ Circular dependencies → resolve() hangs in infinite recursion without detection

04 同仓库更多 Skills

Horonomos-laps
Horonomos-laps/Yana-AI

openai--sentry--sentry

Use when the user asks to inspect Sentry issues or events, summarize recent production errors, or pull basic Sentry health data via the Sentry API; perform read-only queries with the bundled script and require `SENTRY_AUTH_TOKEN`.

★ 13
Horonomos-laps
Horonomos-laps/Yana-AI

ito-basket-compare

Compare Itô prediction-market baskets against a user's knowledge base, portfolio notes, financial context, watchlist, or research thesis. Use for read-only basket comparison and gap analysis without investment advice or live trading.

★ 13
Horonomos-laps
Horonomos-laps/Yana-AI

implementing-network-traffic-baselining

Build network traffic baselines from NetFlow/IPFIX data using Python pandas for statistical analysis, z-score anomaly detection, and hourly/daily traffic pattern profiling

★ 13
Horonomos-laps
Horonomos-laps/Yana-AI

detecting-evasion-techniques-in-endpoint-logs

Detects defense evasion techniques used by adversaries in endpoint logs including log tampering, timestomping, process injection, and security tool disabling. Use when investigating suspicious endpoint behavior, building detection rules for evasion tactics, or conducting threat hunting for stealthy adversary activity. Activates for requests involving evasion detection, defense evasion analysis, log tampering detection, or MITRE ATT&CK TA0005.

★ 13
Horonomos-laps
Horonomos-laps/Yana-AI

implementing-microsegmentation-with-guardicore

Implementing microsegmentation using Akamai Guardicore Segmentation to map application dependencies, create granular network policies, visualize east-west traffic flows, and enforce least-privilege communication between workloads across data centers and cloud.

★ 13
Horonomos-laps
Horonomos-laps/Yana-AI

ramdisk-tmpfs-patterns

暂无简介

★ 13