domains / research / knowledge-graph-construction / knowledge-graph
The two copies are not reproductions of the same procedure. The first is a concrete, self-contained skill for building a knowledge graph from a list of concepts and exporting it to Mermaid. The second is a meta-skill (an index) that routes to nine different knowledge-graph-related skills, including ontology design, citation networks, and specific engines like Graphiti and OpenSPG. Neither contains the material steps of the other, so crowning one would hide a real alternative.
| Repository | Path |
|---|---|
| LeoYeAI/openclaw-master-skills | skills/scholargraph/knowledge-graph/skill.md |
| brycewang-stanford/Auto-Empirical-Research-Skills | skills/43-wentorai-research-plugins/skills/tools/knowledge-graph/SKILL.md |
| Role | Repository | Path |
|---|---|---|
| peer | LeoYeAI/openclaw-master-skills | skills/scholargraph/knowledge-graph/skill.md |
| peer | brycewang-stanford/Auto-Empirical-Research-Skills | skills/43-wentorai-research-plugins/skills/tools/knowledge-graph/SKILL.md |
# Knowledge Graph Builder ## Overview This skill covers two approaches to knowledge graph tasks: a direct builder for concept networks and a meta-index for specialized research graph tools. ## 1. Direct Concept Graph Builder Use this when you have a list of concepts and need to visualize their relationships. ### Steps 1. **Identify Concepts**: Define the list of concepts to include (e.g., `['transformer', 'attention', 'BERT', 'GPT']`). 2. **Build Graph**: Use the `KnowledgeGraph` builder to automatically identify relationships and hierarchy. ```typescript import KnowledgeGraph from './scripts/graph'; const builder = new KnowledgeGraph(); const graph = await builder.build(['transformer', 'attention', 'BERT', 'GPT']); ``` 3. **Export/Visualize**: Convert the graph to a Mermaid diagram for display. ```typescript const mermaidCode = builder.toMermaid(graph); ``` 4. **Navigate**: Use path discovery to find connections between specific nodes. ```typescript const path = builder.findPath(graph, 'attention', 'GPT'); ``` ## 2. Specialized Research Graph Index Use this when the task involves specific research workflows, ontologies, or citation analysis. ### Steps 1. **Select Sub-skill**: Choose the appropriate tool from the following list based on the user's need: - **Citation Analysis**: `citation-network-builder` - **Concept Mapping**: `concept-map-generator` - **AI Agent Graphs**: `graphiti-guide` - **RAG Systems**: `knowledge-graph-construction` or `rag-methodology-guide` - **Ontology Design**: `ontology-design-guide` - **Specific Engines**: `openspg-guide` (Ant Group SPG/KAG) - **Reference Syncing**: `notero-zotero-notion-guide` or `zotero-markdb-connect-guide` 2. **Read Documentation**: `read` the `SKILL.md` of the selected sub-skill to follow its specific procedure. ## Conflicts - **brycewang-stanford/Auto-Empirical-Research-Skills**: This repository treats 'knowledge-graph' as a category of 9 distinct skills, requiring a selection step before any construction can occur. It does not provide a direct builder script. - **LeoYeAI/openclaw-master-skills**: This repository provides a direct, code-based builder for concept graphs without referencing external sub-skills or ontology design tools.