ladybugdb-expert

name	ladybugdb-expert
version	1.0.0
description	Comprehensive knowledge of LadybugDB, the active community fork of the archived Kuzu graph database. Covers the Python SDK (real_ladybug package), Cypher query language, vector search extensions, full-text search, graph algorithms, import/export, and migration from Kuzu.
triggers	["ladybugdb"]

Ladybugdb Expert Skill

Knowledge graph: 43 articles, 207 entities, 178 relationships

Overview

Comprehensive knowledge of LadybugDB, the active community fork of the archived Kuzu graph database. Covers the Python SDK (real_ladybug package), Cypher query language, vector search extensions, full-text search, graph algorithms, import/export, and migration from Kuzu.

Usage

This skill provides deep domain expertise through a knowledge graph. Ask questions naturally and the skill will retrieve relevant information from the graph database.

Example queries:

• "Explain ladybugdb concepts"
• "What is the relationship between X and Y in ladybugdb?"
• "Find articles about [topic] in ladybugdb"

Knowledge Domains

This pack covers 43 articles across multiple domains within ladybugdb.

Quality Metrics:

• Accuracy: 0.0%
• Hallucination Rate: 0.0%
• Citation Quality: 0.0%

Technical Details

• Database: Kuzu graph database (2.08 MB)
• Configuration: data/packs/ladybugdb-expert/kg_config.json
• License: MIT

Sources

This knowledge pack was created from:

• https://docs.ladybugdb.com/
• https://docs.ladybugdb.com/client-apis/python/
• https://docs.ladybugdb.com/extensions/vector/
• https://github.com/LadybugDB/ladybug

Integration

This skill integrates with WikiGR's KG Agent for semantic retrieval.

How to Use This Pack:

When a user asks a question related to this domain, automatically:

from wikigr.agent.kg_agent import KnowledgeGraphAgent
from pathlib import Path

# Initialize KG Agent with pack database
pack_db = Path("data/packs/ladybugdb-expert/pack.db").resolve()
with KnowledgeGraphAgent(db_path=str(pack_db), read_only=True) as kg_agent:
    result = kg_agent.query(user_question, max_results=5)
    # Use result["answer"] and result["sources"] in your response

Retrieval Strategy:

1. Vector Search: Find semantically similar content
2. Graph Traversal: Navigate relationships between entities
3. Hybrid Search: Combine vector similarity with graph structure

Created: 2026-03-03T18:44:21.748608Z