AWS Adds Managed Web Search Connector for Bedrock AgentCore
AWS says AgentCore can now expose web search as an MCP-compatible managed target, moving fresh public-web retrieval behind the same Gateway controls enterprises already use for agents.
Illustrative image. Cedar S. Insights uses editorial stock photography; images do not depict specific events described in articles.
AWS announced Web Search on Amazon Bedrock AgentCore on 19 June 2026, positioning it as a managed way for enterprise agents to retrieve current public-web information through AgentCore Gateway rather than through separately provisioned search APIs.
The AWS blog says the capability is generally available and MCP-compatible. Developers attach it to a Gateway with the connector ID web-search; agents can discover it through the standard tools/list flow and invoke it like other MCP tools.
AWS documentation says the tool returns semantically relevant snippets, URLs, titles and publication dates in text and structured JSON formats. The company says query traffic stays inside AWS infrastructure, is not sent to a third-party search engine, and is not used to train models or improve the service.
The announcement says the connector is available in us-east-1 and prices web-search use at $7 per 1,000 queries, separate from any surrounding AgentCore or model costs.
Why It Matters
Web-grounded agents are attractive for research, support and analyst workflows, but enterprise rollouts often stall on search keys, data egress, retention policies and result parsing. AWS is turning that plumbing into a managed Gateway target, which could make retrieval-augmented agent designs easier to govern.
Sourcing note: Confirmed facts are the AWS announcement, documentation, connector name, region statement and published pricing. Claims about index scale, freshness, privacy handling and retrieval quality are AWS statements and should be validated in customer pilots.
Why It Matters
If the managed connector performs as described, AWS is reducing one of the messier parts of production AI agents: safely giving models fresh public information without routing enterprise questions through external search services. The claims still need workload-specific validation.
Primary Sources
Our sourcing: Cedar S. Insights provides source-led editorial analysis. Reported company, institutional and regulatory claims are attributed to their original sources unless stated otherwise.
Corrections: If a material factual error is identified, Cedar S. Insights will update the relevant article and preserve the distinction between the corrected statement and supporting evidence.
Topics