Oriel — AI & technology worth noticing, every day.

Introducing the Web Search API on Amazon Bedrock AgentCore

Web page · https://aws.amazon.com/blogs/machine-learning/introducing-the-web-search-api-on-amazon-bedrock-agentcore

Today we're announcing the Web Search API on Amazon Bedrock AgentCore: your own code can search the web with a single signed API call, with no gateway and no resources to create. Call it from the AWS SDKs or the AWS CLI and use it in applications, agent loops, and data pipelines.

First seen by Oriel on .

AI summary

Amazon Bedrock AgentCore’s Web Search API lets applications query Amazon’s web index directly through one IAM-authenticated API call, without creating a gateway or managing API keys.

  • InvokeWebSearch returns up to 25 ranked results, each containing a text extract and optionally a title, URL, and publication time.
  • Requests support inclusive RFC 3339 publication-date ranges and include or exclude lists of up to 100 domains, with root domains matching subdomains.
  • The API uses SigV4, requires bedrock-agentcore:InvokeWebSearch permission, supports AWS SDKs and AWS CLI, and defaults to 10 requests per second per account and Region.
  • LangChain support requires langchain-aws 1.9.3 or later, while AgentCore Python SDK integration requires version 1.26.0 or later.
  • AWS PrivateLink enables access from private subnets without internet, NAT, or proxy connectivity, with endpoint policies and aws:SourceVpce conditions providing additional controls.

#amazon bedrock #web search #llm agents #aws privatelink

Generated by gpt-5.6-sol. Check the original for details.

Aliases

SourceExternal IDURL
AWS Machine Learning Blog811b8dfa171613df698db45e0ea693304815500bhttps://aws.amazon.com/blogs/machine-learning/introducing-the-web-search-api-on-amazon-bedrock-agentcore/ Read discussion