Introducing Web Search on Amazon Bedrock AgentCore
Amazon Bedrock AgentCore now offers a fully managed web search capability, allowing AI agents to access up-to-date information from the web without infrastructure overhead. This feature, compatible with the Model Context Protocol (MCP), provides a purpose-built web index spanning tens of billions of documents, updated continually to reflect new content within minutes. The privacy model ensures that queries stay within AWS, and retrieval can combine a knowledge graph with semantic snippet extraction. This development has significant implications for engineers building AI systems, as it addresses the limitation of frozen knowledge at training time and enables agents to respond to real-time queries.
⚡ Key Takeaways
- Web Search on Amazon Bedrock AgentCore is a fully managed, MCP-compatible web search capability.
- The feature is backed by a purpose-built web index operated by Amazon, spanning tens of billions of documents.
- The index is updated continually, reflecting new content within minutes, to provide high-confidence responses to factual questions.
- The knowledge graph provides high-confidence responses for factual questions, reducing the need for model inference.
- The privacy model ensures that queries stay within AWS, addressing concerns about data retention and reuse.
The introduction of Web Search on Amazon Bedrock AgentCore has significant implications for engineers building AI systems, as it enables agents to access up-to-date information from the web and respond to real-time queries. This development addresses the limitation of frozen knowledge at training time, allowing agents to provide more accurate and relevant responses.
✅ Practical Steps
- Connect to the AgentCore Gateway using AWS Identity and Access Management (IAM) or JSON Web Token (JWT) inbound auth.
- Discover the Web Search capability with a standard tools/list call and invoke it like other MCP tools.
- Use the managed connector to route queries to the Web Search tool in the AWS service account, without provisioning search APIs or managing outbound credentials.
Want the full story? Read the original article.
Read on AWS ML Blog ↗