npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@pipedog/aws

v0.11.0

Published

AWS services (Bedrock LLM, Transcribe STT, Polly TTS, Nova Sonic, AgentCore, SageMaker) for pipedog.

Downloads

1,609

Readme

@pipedog/aws

AWS services for pipedog:

  • AWSBedrockLLMService: Bedrock models through the Converse API, with AWSBedrockLLMAdapter.
  • AWSTranscribeSTTService: streaming speech-to-text from Amazon Transcribe.
  • AWSPollyTTSService: text-to-speech from Amazon Polly.
  • AWSNovaSonicLLMService: Amazon Nova Sonic speech-to-speech.
  • AWSAgentCoreProcessor: an agent on Bedrock AgentCore Runtime, used in place of an LLM.
  • SageMakerBidiClient: bidirectional streaming to models on SageMaker endpoints.

Most of them call AWS with fetch and WebSocket, signing requests with SigV4 on WebCrypto, so no AWS SDK is needed.

A port of Pipecat's pipecat.services.aws, pipecat.adapters.llm.services.bedrock_adapter and pipecat.adapters.llm.services.aws_nova_sonic_adapter. Subpaths mirror upstream: @pipedog/aws/llm, /stt, /tts, /agent_core, /nova_sonic/llm, /sagemaker/bidi_client, /bedrock_adapter, /aws_nova_sonic_adapter and /utils.

Install

bun add @pipedog/core @pipedog/asyncio @pipedog/aws

Two classes can use an AWS SDK client, both optional peer dependencies:

bun add @aws-sdk/client-sagemaker-runtime-http2  # SageMakerBidiClient (always)
bun add @aws-sdk/client-bedrock-runtime          # AWSNovaSonicLLMService with transport: "http2"

Usage

import { AWSBedrockLLMService } from "@pipedog/aws/llm";
import { AWSTranscribeSTTService } from "@pipedog/aws/stt";
import { AWSPollyTTSService } from "@pipedog/aws/tts";

// Keys come from the options or AWS_ACCESS_KEY_ID / AWS_SECRET_ACCESS_KEY /
// AWS_SESSION_TOKEN / AWS_REGION.
const stt = new AWSTranscribeSTTService({ region: "us-east-1" });
const llm = new AWSBedrockLLMService({
  awsRegion: "us-east-1",
  settings: new AWSBedrockLLMService.Settings({
    model: "us.amazon.nova-lite-v1:0",
    systemInstruction: "You are a helpful voice assistant. Keep replies short.",
  }),
});
const tts = new AWSPollyTTSService({ region: "us-east-1" });
// new Pipeline([transport.input(), stt, userAggregator, llm, tts, transport.output(), ...])

Credentials

Keys are read, in order, from the service's options, the environment, and a credentialsProvider function. The provider is called for every request or connection, so it can renew temporary credentials, for example fromNodeProviderChain() from @aws-sdk/credential-providers (Node only). There is no built-in default credential chain (profiles, instance metadata, ECS roles, SSO).

Runtimes

| | Bun | Node ≥ 22 | Deno 2 | Cloudflare Workers | |---|---|---|---|---| | AWSBedrockLLMService | ✅ | ✅ | ✅ | ✅ | | AWSTranscribeSTTService | ✅ | ✅ | ✅ | ✅ | | AWSPollyTTSService | ✅ | ✅ | ✅ | ✅ | | AWSAgentCoreProcessor | ✅ | ✅ | ✅ | ✅ | | AWSNovaSonicLLMService (WebSocket, default) | ✅ | ✅ | ✅ | ✅ | | AWSNovaSonicLLMService (transport: "http2") | ✅ | ✅ | untested | ❌ (HTTP/2 duplex) | | SageMakerBidiClient | ✅ | ✅ | untested | ❌ (HTTP/2 duplex) |

Nova Sonic uses Bedrock's WebSocket endpoint for InvokeModelWithBidirectionalStream (the one the AWS SDK uses in browsers), with SigV4-presigned URLs and signed events, so it runs everywhere. SageMaker's InvokeEndpointWithBidirectionalStream has no WebSocket endpoint and needs a full-duplex HTTP/2 stream, which fetch cannot provide and Workers do not offer. The SDK clients are loaded lazily, so importing the package works on Workers. Both classes accept an injected client, so you can supply another transport.

compat/smoke.ts runs Bedrock, Polly and AgentCore against in-process fake fetch functions on every runtime, checks SigV4 against an AWS test vector, and constructs the other classes.

Links