The Curated Data blog
The answers behind your AI, explained
How approved answers work, how the review workflow keeps unapproved edits away from your assistants, and practical guides for connecting the AI tools your team already uses.
- Published 7 min read
RAG retrieves. Review decides what your agent should trust.
Vector search can find relevant content without knowing which version is approved, current, or allowed. Here is how retrieval and curation fit together in a production agent.
RAGarchitectureRead → - Published 8 min read
What belongs in an AI agent knowledge base?
A practical inclusion test, six high-value knowledge domains, a copyable concept template, and a first-week plan for building an approved source without importing every file.
knowledge operationsguideRead → - Published 7 min read
Why AI assistants guess, and how to give yours approved answers instead
When an assistant cannot reach the answer your company approved, it fills the gap from training data. Here is the mechanism that closes that gap: approved pages, traceable revisions, connected knowledge, and retrieval at answer time.
groundingagentsRead → - Published 6 min read
Introducing Curated Data: write, approve, serve
Curated Data turns what your company knows into answers your team approved. A tour of the workflow, from a first page to an assistant answering from the approved version.
announcementproductRead → - Published 8 min read
Connecting your knowledge to agents over MCP
A practical guide to wiring an MCP-capable agent to your approved knowledge base, with the real tool calls, REST requests, and responses the live API returns.
MCPguideRead → - Published 6 min read
The review workflow: the approval gate before anything reaches an assistant
How the draft → in-review → approved lifecycle works, who can approve, and why the self-approval setting is a deliberate choice, so nothing your agents read was skipped past your team.
reviewgovernanceRead → - Published 6 min read
Versioned knowledge without whole-workspace publishing
Approval publishes one document immediately while preserving its numbered revision and review history. Here’s why that model keeps governance strong without a bundle release step.
versioningreviewRead →
