MCP Knowledge Graph

An architectural pattern in which an enterprise **knowledge graph** (Neo4j, PuppyGraph, TigerGraph, ArangoDB, or a custom triple store) is exposed to LLM agents through an MCP server, allowing agents to traverse multi-hop graph relationships using standardized MCP `resources` and `tools` primitives instead of generating brittle Cypher / Gremlin / SPARQL queries client-side. The MCP server translates the agent's natural-language intent into safe, parameterized graph traversals and returns structured node-edge results to the agent's context window.

Vector-similarity search (embeddings + ANN over a vector database) handles unstructured documents well but degrades sharply on questions that require *connecting* multiple known entities ("which suppliers of company X are also suppliers of company Y's competitors?"). Knowledge graphs answer these directly via multi-hop traversal — but writing the traversal query correctly is a hard problem for LLMs (Cypher generation is hallucination-prone). MCP Knowledge Graph servers solve this by exposing pre-defined, parameterized traversal *tools* (e.g., `find_path_between(node_a, node_b, max_hops=N)`) rather than asking the model to generate raw query language. The graph stays canonical; the agent stays grounded.