tigergraph

What This Skill Does

The tigergraph skill provides a comprehensive interface for interacting with TigerGraph, a high-performance distributed graph database. It enables you to perform deep-link analysis across massive, interconnected datasets that traditional relational databases struggle to process. By utilizing GSQL, a powerful graph query language, this skill allows for parallel traversal of billions of edges with sub-second latency. Whether you are modeling complex relationships for fraud detection, supply chain management, or enterprise knowledge graphs, this skill serves as your authoritative reference for schema definition, REST API integration, and GSQL query implementation.

Installation

To integrate the TigerGraph documentation into your OpenClaw environment, execute the following command in your terminal: clawhub install openclaw/skills/skills/bytesagain/tigergraph Once installed, you can access the full technical reference, command documentation, and best practices directly within your agent workflow.

Use Cases

• Fraud Detection: Rapidly traverse multi-hop transaction networks to uncover circular money laundering patterns or suspicious actor rings in real time.
• Customer 360: Aggregate fragmented data from disparate business units into a unified graph entity, providing a holistic view of user behavior and influence.
• Supply Chain Optimization: Model complex, multi-tiered supplier dependencies to identify critical bottleneck nodes and mitigate risks before they propagate.
• Knowledge Graphs: Build and query expansive enterprise knowledge bases, allowing for semantic discovery across structured and unstructured data points.
• Recommendation Engines: Utilize built-in graph algorithms like PageRank or collaborative filtering to deliver hyper-personalized suggestions based on network topology.

Example Prompts

1. "Tigergraph, explain the syntax for creating an undirected edge between two vertex types in GSQL and how to add an attribute to that edge."
2. "I need to run a PageRank algorithm on a graph with 50 million edges. What is the most efficient way to call the pre-installed PageRank query via the REST API?"
3. "Compare TigerGraph's parallel processing architecture to Neo4j. When would I prefer TigerGraph's MPP approach for a deep-link retail analytics project?"

Tips & Limitations

To maximize the effectiveness of this skill, ensure your graph schema is optimized for the depth of your traversals. Remember that while GSQL is powerful, complex multi-hop queries should be written as installed queries rather than ad-hoc queries to leverage compiled performance. This skill is strictly a reference tool; it does not execute code against an external database instance. Always verify your query logic against small, representative subsets of your data before running analytics on production-scale clusters.