googlecloudplatform/knowledge-catalog

Google's Knowledge Catalog is a tool that helps organizations make sense of all their data, documents, databases, spreadsheets, and more, by organizing it into a smart, searchable map. Instead of data sitting in scattered silos where no one can find it or understand what it means, the catalog adds context so that both people and AI systems can actually use that data effectively. This repository collects tools, agents, and code samples that show how to work with that platform. At a high level, the catalog builds what's called a "knowledge graph", think of it as a web of connections between your data and the meaning behind it. For example, if your company has a spreadsheet called "Q3 Revenue," the catalog can link it to related documents, define what "revenue" means in your business context, and make that information available to AI assistants. The samples in this repo demonstrate how to build solutions around context management, enrichment, and retrieval, essentially, how to help AI agents understand your data well enough to answer questions about it accurately. This is aimed at teams already working within Google Cloud who need to manage large amounts of enterprise data and want to make it useful for AI applications. A practical example: a company with thousands of internal policies, reports, and datasets could use these tools to build an AI assistant that actually understands which documents matter and what terms mean in that specific organization, rather than guessing. The README is quite sparse on technical specifics. It offers a button to open everything in Google Cloud Shell, which lets you explore the code directly in a browser-based environment without setting anything up locally. Beyond that, the README doesn't go into detail about individual tools or how to configure them, you'd need to explore the repository contents directly. The project is open source under the Apache 2.0 license, though the repo notes it isn't an officially supported Google product.