ai-infrastructure-hugg…
Hugging Face Inference SDK patterns for TypeScript/Node.js — InferenceClient setup, chat completion, text generation, streaming, embeddings, image generation,…
AWS SDK v3 for TypeScript — modular clients, command pattern, S3, DynamoDB, SQS, Lambda, SNS, Secrets Manager
$ npx -y skills add agents-inc/skills --skill infra-platform-aws-sdk --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/infra-platform-aws-sdkContext preview
The summary Claude sees to decide when to auto-load this skill.
AWS SDK v3 for TypeScript — modular clients, command pattern, S3, DynamoDB, SQS, Lambda, SNS, Secrets Manager
name: infra-platform-aws-sdk description: AWS SDK v3 for TypeScript — modular clients, command pattern, S3, DynamoDB, SQS, Lambda, SNS, Secrets Manager
> **Quick Guide:** AWS SDK v3 for JavaScript/TypeScript uses modular packages (`@aws-sdk/client-*`) with a command pattern: create a client, instantiate a command, call `client.send(command)`. Import only the services you need for tree-shaking. Use `DynamoDBDocumentClient` for native JS types. Use `getSignedUrl` from `@aws-sdk/s3-request-presigner` for presigned URLs. Handle errors with `instanceof` specific exception classes. Use built-in paginators (`paginate*`) with `for await...of`.
---
<critical_requirements>
> **All code must follow project conventions in CLAUDE.md** (kebab-case, named exports, import ordering, `import type`, named constants)
**(You MUST use AWS SDK v3 modular packages (`@aws-sdk/client-*`) — NEVER the monolithic `aws-sdk` v2 package)**
**(You MUST use the command pattern: `client.send(new XxxCommand({...}))` — NEVER call methods directly on the client)**
**(You MUST use `DynamoDBDocumentClient` from `@aws-sdk/lib-dynamodb` for DynamoDB — it auto-marshalls native JS types)**
**(You MUST handle errors with `instanceof` specific exception classes — NEVER catch generic `Error` and check `.code`)**
**(You MUST use built-in paginators (`paginate*` functions) for paginated APIs — NEVER manually track continuation tokens)**
</critical_requirements>
---
---
**Auto-detection:** AWS SDK, @aws-sdk/client, S3Client, DynamoDBClient, DynamoDBDocumentClient, SQSClient, LambdaClient, SNSClient, SecretsManagerClient, PutObjectCommand, GetObjectCommand, GetCommand, PutCommand, QueryCommand, SendMessageCommand, InvokeCommand, GetSecretValueCommand, getSignedUrl, s3-request-presigner, credential-providers, fromEnv, fromIni, paginateListObjectsV2, aws-sdk-client-mock
**When to use:**
**When NOT to use:**
**Key patterns covered:**
---
<philosophy>
AWS SDK v3 is a ground-up rewrite of the v2 SDK for modern JavaScript/TypeScript. The core design principles:
1. **Modular packages** — Each service is a separate npm package (`@aws-sdk/client-s3`, `@aws-sdk/client-dynamodb`). Import only what you use. This reduces bundle size by up to 90% compared to the monolithic v2 `aws-sdk` package.
2. **Command pattern** — Every API call is a Command object sent through a Client. This enables middleware, type safety, and testability. The client handles serialization, signing, retries, and deserialization.
3. **First-class TypeScript** — Every command input and output is fully typed. Use the types to avoid runtime errors.
4. **Middleware stack** — Customize request/response handling at various stages (serialize, build, finalize, deserialize) without monkey-patching.
5. **Built-in pagination** — Paginator functions return async iterators, eliminating manual token tracking.
**When to use AWS SDK v3:**
**When NOT to use:**
</philosophy>
---
<patterns>
Every AWS service follows the same pattern: import the client and command, create a client instance, send the command.
import { S3Client, PutObjectCommand } from "@aws-sdk/client-s3";
const s3 = new S3Client({ region: "us-east-1" });
await s3.send(
new PutObjectCommand({
Bucket: "my-bucket",
Key: "data.json",
Body: JSON.stringify({ hello: "world" }),
ContentType: "application/json",
}),
);**Why good:** modular import keeps bundle small, command pattern enables middleware and type safety, region is explicit
Create clients once and reuse them — they manage connection pooling internally. In Lambda, create clients outside the handler for connection reuse across invocations.
See [examples/core.md](examples/core.md) for cl
The official skills marketplace for Agents Inc. 150+ skills covering everything from React and Prisma to Redis, ElevenLabs, and infrastructure tooling. Pick the skills that match your stack and install them via Claude Code. Need more control?
Repo: agents-inc/skills
Hugging Face Inference SDK patterns for TypeScript/Node.js — InferenceClient setup, chat completion, text generation, streaming, embeddings, image generation,…
LiteLLM proxy server setup, TypeScript client patterns via OpenAI SDK, model routing, fallbacks, load balancing, spend tracking, virtual keys, and production…
Serverless GPU compute platform for AI model deployment — web endpoints, GPU functions, model serving, and TypeScript client patterns
Local LLM inference with the Ollama JavaScript client -- chat, streaming, tool calling, vision, embeddings, structured output, model management, and…
Replicate SDK patterns for TypeScript/Node.js -- client setup, predictions, streaming, webhooks, file handling, model versioning, deployments, and training
Together AI SDK patterns for TypeScript — client setup, chat completions, streaming, structured output, function calling, embeddings, image generation,…