Build AI agents that understand data meaning and context
Accumulated data is difficult to utilize if AI does not understand its meaning. This repository demonstrates how to convert data structures and business context into knowledge graphs that AI agents can understand. You can build data retrieval and augmentation solutions using the Google Cloud Knowledge Catalog feature.
Knowledge Catalog for data meaning
Knowledge Catalog, formerly known as DataFlex, is an AI-based data catalog and metadata management platform. It provides a dynamic knowledge graph for all structured and unstructured data. This allows AI agents to grasp not only the data itself but also the underlying meaning and business context. This repository collects tools, agents, and sample code that demonstrate these features.
Building context management and retrieval solutions
The samples provided in this repository show the process of building data context management, augmentation, and retrieval solutions. Developers can use these examples to design AI agent workflows suitable for their data environments. You can see how to structure the necessary metadata so that AI agents can correctly interpret and utilize data. The code, written in TypeScript, helps in understanding the actual implementation methods.
Things to check before use
This repository is not an official Google product, and all solutions are provided under the Apache 2.0 license. Therefore, you must review the license terms before adopting them in an enterprise environment. Additionally, since the Knowledge Catalog service itself runs on Google Cloud infrastructure, it is advisable to understand the access permissions and cost structure of that cloud environment in advance. The sample code is provided for reference, so it is safe to undergo internal testing and security verification before applying it to a production environment.