
Understand Knowledge
Turns a Karpathy-pattern LLM wiki (the three-layer setup with raw sources, wikilinked markdown, and a schema file) into an interactive knowledge graph. It scans your wiki directory for articles and wikilinks, then dispatches subagents to extract implicit relationships and entities that aren't explicitly linked. The output is a force-directed graph with topic clustering and a guided tour built from your index.md structure. Useful if you're maintaining one of these wikis and want to surface the conceptual structure that's hiding between the explicit wikilinks. The deterministic extraction is all Python, LLM agents only handle inference work.
- skills.sh installs
- 3.9K
- GitHub stars
- 91.8K
- last updated
- Aug 21, 2026
Install
$ npx -y skills add egonex-ai/understand-anything --skill understand-knowledge --agent claude-codeInstalls into .claude/skills of the current project.
Files
/understand-knowledge
Analyzes a Karpathy-pattern LLM wiki — a three-layer knowledge base with raw sources, wiki markdown, and a schema file — and produces an interactive knowledge graph dashboard.
What It Detects
The Karpathy LLM wiki pattern (see https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f):
- Raw sources — immutable source documents (articles, papers, data files)
- Wiki — LLM-generated markdown files with wikilinks (
[[target]]syntax) - Schema — CLAUDE.md, AGENTS.md, or similar configuration file
- index.md — content catalog organized by categories
- log.md — chronological operation log
Detection signals: has index.md + multiple .md files with wikilinks. May have raw/ directory and schema file.
Instructions
Phase 1: DETECT
Sponsors
Details
- Repository
- github.com/egonex-ai/understand-anything
- Categories
- Documentation, AI & Agent Building
- First seen
- Jun 11, 2026