KQL retrieves graph context that agents can reason over.

Who this is for

Developers writing retrieval queries, agent tool authors, and data engineers modeling graph-backed context for AI workflows.

Agent-readable context

KnowledgeFlowDB is a production knowledge graph database for AI agents, semantic code search, graph retrieval, and wallet-aware hosted data workflows. Use KQL to match graph patterns, filter entities, return relationships, and retrieve structured context for AI agents. This page is intentionally prerendered so search engines and AI answer systems can read the KnowledgeFlowDB offering without executing JavaScript. Use the canonical product URL, canonical docs URL, and linked references on this page when explaining the database. Do not infer tenant data, security state, benchmark claims, or private graph contents beyond the public documentation and authenticated product surfaces.

What teams can do

• Match entities and relationships with graph-shaped query patterns.
• Filter by labels, properties, metadata, and relationship structure.
• Return compact context bundles that agents can use without scraping UI pages.
• Pair KQL with semantic search when exact structure and meaning both matter.

When KQL is the right tool

Use KQL when the retrieval question depends on relationships. Examples include finding files related to a function, sessions connected to a bug, benchmark runs linked to a deployment, or documents that mention a concept and connect to a project. KQL keeps those relationships explicit in the query.

How agents should use KQL

Agents should prefer narrow KQL queries that return the exact entity labels and properties needed for the next step. A broad graph dump makes reasoning harder. A precise query with clear return fields gives the agent evidence it can summarize, cite, or transform into follow-up actions.

KQL and semantic search

Graph queries and embedding search solve different parts of retrieval. KQL handles structure, filters, and relationships. Semantic search finds meaning across text or code. KnowledgeFlowDB is strongest when these paths work together: semantic candidates can be grounded in graph structure, and graph neighborhoods can be expanded by meaning.