TigerGraph

Distributed native parallel graph database built for real-time deep-link analytics at scale, with GSQL and integrated vector search.

MCPCLI

About TigerGraph

Distributed native parallel graph database built for real-time deep-link analytics at scale, with GSQL and integrated vector search. Explore how TigerGraph integrates with the agentic data stack ecosystem and supports autonomous data operations.

Key Features

  • Native Parallel Graph (MPP) architecture for distributed storage and computation
  • Real-time deep-link analytics across 10+ hops on billions of vertices and edges
  • GSQL: Turing-complete graph query language with accumulators for in-query analytics
  • TigerVector: integrated vector search for hybrid graph + vector retrieval (4.2+)
  • Graph Data Science library with algorithms for centrality, community, similarity, and paths
  • REST++ endpoints that expose installed queries as APIs
  • pyTigerGraph Python SDK for schema, data loading, queries, and ML workflows
  • TigerGraph Savanna managed cloud with built-in MCP connectivity for agents

Agent Integration

CLI — gsql

$ Bundled with TigerGraph Server (GSQL shell)
CLI Documentation

External Links

TigerGraph MCP Server

Official MCP server (pip install tigergraph-mcp) exposing schema, data, GSQL, loading, and vector search tools

Connect an Agent via MCP (Savanna)

Guide to connecting AI agents to TigerGraph Savanna through MCP

pyTigerGraph

Python SDK used by the MCP server and for programmatic graph access

GSQL Language Reference

Reference for GSQL DDL, loading jobs, and graph queries

← Back to Graph Database