knowledge-graph-construction — Claude Code skill: setup guide
knowledge-graph-construction guides the design of knowledge graphs from unstructured data, covering model selection between LPG and RDF, schema creation, and extraction for RAG.
no license 103 installs ★ 203 updated 2026-09-01
npx skills add lyndonkl/claude --skill knowledge-graph-construction -a claude-codeWhat knowledge-graph-construction does
The skill walks through a six-stage checklist to turn raw documents into a graph. It helps select the core data model: Labeled Property Graphs for RAG pipelines or RDF/OWL for ontologies like UMLS.
The architecture separates data into three distinct tiers: document instances, curated domain knowledge, and formal ontologies. Relation extraction combines targeted prompts with a second LLM pass to spot missed facts. Graph validation inspects connectivity and checks for orphan nodes.
Standard LPG stores lack full logical reasoning. Projects that demand formal inference will fail unless built on RDF/OWL.
What it helps with
- choose between LPG, RDF, and hypergraph formats
- partition entities and relationships across trust layers
- configure text extraction pipelines with dual-pass verification
- detect contradictory edges and orphan nodes during quality checks
Example prompts for the agent
- Help me choose between LPG and RDF for an enterprise RAG setup.
- Design an event reification schema for multi-party clinical trials.
- Build a two-pass extraction pipeline to extract relations from research papers.
How to install knowledge-graph-construction
1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill knowledge-graph-construction -a claude-code2Manually: put the files in place
~/.claude/skills/knowledge-graph-construction/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/knowledge-graph-construction/
1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill knowledge-graph-construction -a codex2Manually: put the files in place
.agents/skills/knowledge-graph-construction/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill knowledge-graph-construction -a cursor2Manually: put the files in place
.cursor/skills/knowledge-graph-construction/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install lyndonkl/claude skills/knowledge-graph-construction2Manually: put the files in place
.github/skills/knowledge-graph-construction/SKILL.md1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill knowledge-graph-construction -a gemini-cli2Platform CLI
gemini skills install https://github.com/lyndonkl/claude --path skills/knowledge-graph-constructionManually: put the files in place
.devin/skills/knowledge-graph-construction/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill knowledge-graph-construction -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about knowledge-graph-construction
How do I install knowledge-graph-construction in Claude Code?
Run npx skills add lyndonkl/claude --skill knowledge-graph-construction -a claude-code.
What does knowledge-graph-construction do?
It guides the creation of knowledge graphs, covering storage model selection, schema design, and text extraction pipelines.
When should RDF/OWL be chosen instead of LPG?
Choose RDF/OWL when working with established ontologies such as UMLS or SNOMED, or when the system demands formal OWL-DL reasoning.