Elastic has launched Agent Builder, a developer tool for creating custom AI agents that utilize proprietary data. This tool integrates seamlessly with Elasticsearch, enhancing context engineering by ensuring AI agents access the right data at the right time. It addresses challenges related to scattered and unstructured data, offering a unified platform for agent development, execution, and monitoring.
Agent Builder features a native conversational agent, allowing users to query company data in natural language instead of structured queries, improving the relevance of AI outputs while ensuring security and compliance. Developers can customize tools using Elasticsearch’s ES|QL for precise access control and can build bespoke agents catered to organizational needs.
Key features include real-time chat functionality, intelligent search capabilities, and safe integration with external applications through Model Context Protocol (MCP) and Agent-to-Agent (A2A) connections. Currently in Technical Preview on Elastic Cloud, Agent Builder will soon be available in version 9.2.
Source link
