Google Data Agent Kit Adds Direct Cloud Service Control
The open-source framework enables AI agents to interface with Google Data Cloud services using standard security policies.
Updated on Oct. 24, 2026 in Artificial Intelligence
Live Poll
Do you trust AI agents to manage your company's data access automatically?
Google has moved its Data Agent Kit to general availability, providing an open-source framework designed to connect AI coding agents to more than 15 Google Data Cloud services. This toolset enables agents to interpret data structures and interact directly with cloud resources using established security protocols.
Why it matters
The kit bridges the gap between generic AI coding assistants and specific data environments by giving agents context about table schemas and partitioning. It standardizes how models interact with enterprise data while enforcing security through explicit access controls.
The framework utilizes Model Context Protocol (MCP) servers to expose live schema and tools, supporting 15 Google Data Cloud services. It includes 37 skills and enforces security via read-only service accounts or IAM deny policies.
The players
Google Cloud
A suite of cloud computing services that provides the infrastructure, data storage, and analytics platforms required for enterprise AI deployment.
The details
The Data Agent Kit functions as an intermediary, sitting between coding agents and Google Data Cloud services as an MCP server—a standard for connecting AI assistants to data sources. By exposing live schema information and specialized tools, the kit allows agents to perform operations on services like BigQuery and Bigtable. Security is managed through IAM (Identity and Access Management) deny policies that block specific tools, while data loss prevention guardrails require user consent for irreversible actions.
Timeline
September 29, 2026: Data Agent Kit version 1.0.0 was released.
October 13, 2026: Data Agent Kit version 1.1.0 was released.
October 24, 2026: Google announced general availability of the Data Agent Kit.
The Tech Race
This release represents a shift toward standardizing agent-data interactions via the Model Context Protocol, mirroring broader efforts to make AI coding assistants cloud-native. It marks a departure from proprietary silos by offering an open-source framework for Google's specific cloud schema.
Developers can now use the kit within environments like VS Code, Cursor, and Claude Code, as it comes preinstalled in Cloud Shell and Cloud Workstations. Teams should audit their existing IAM service account permissions to ensure these agents operate with the intended read-only access.
The takeaway
The Data Agent Kit provides a necessary technical foundation for agents to interact reliably with complex cloud data structures rather than hallucinating schemas. Practitioners should track future updates to the repository's 37 skills to see which new cloud services receive integration support.
Further reading
Explore more developments in the Artificial Intelligence sector.
Source note: This article includes information reported by InfoQ.
Live Poll
Do you trust AI agents to manage your company's data access automatically?
