Neo4j

What This Skill Does

The Neo4j skill enables the OpenClaw AI agent to interface directly with Neo4j graph databases by generating, optimizing, and executing complex Cypher queries. It is designed to bridge the gap between natural language requests and graph database internals, ensuring that operations like merging, path traversal, and batch processing are performed efficiently while adhering to graph modeling best practices.

Installation

To install this skill, run the following command in your terminal or OpenClaw interface: clawhub install openclaw/skills/skills/ivangdavila/neo4j

Use Cases

This skill is ideal for developers managing highly interconnected data. Use cases include mapping social networks, managing recommendation engines, tracking supply chain relationships, or navigating hierarchical organizational structures. It is particularly powerful when you need to perform high-performance graph algorithms or complex pattern matching that would be inefficient in a traditional relational database.

Example Prompts

1. "Find all users connected to user '123' within 3 degrees, but exclude the starting node and ensure the path only includes active relationships."
2. "Write a Cypher query to perform a batch upsert of 5,000 product nodes from a JSON list provided as a parameter, including unique constraints to prevent duplication."
3. "Analyze the following query and optimize it to avoid a Cartesian product issue: MATCH (a:User), (b:Project) WHERE a.team = b.team RETURN a, b."

Tips & Limitations

To master this skill, pay close attention to performance overhead. Always prefer Unique Constraints over plain indexes when dealing with identifier properties, as they serve both purposes simultaneously. When working with path traversals, strictly define your bounds (e.g., [*1..5]) to prevent the query from exploding on deeply connected nodes. Remember that MERGE can lead to unintended duplicates if you merge an entire path at once; always separate node merges from relationship merges for robust upserts. When managing large datasets, leverage CALL {} IN TRANSACTIONS to avoid exhausting database heap memory. Finally, always use the PROFILE command before running queries in production to check for expensive operations like Cartesian products or full label scans.