understand-knowledge
understand-knowledge processes three-layer knowledge bases (raw sources, wiki markdown, schema) to extract entities, resolve wikilinks, and cluster topics into an interactive graph. It detects the Karpathy wiki pattern, dispatches LLM agents to find implicit relationships, merges results with deterministic extraction, and outputs a navigable knowledge graph dashboard.
understand-knowledge analyzes a Karpathy-pattern LLM wiki and generates an interactive knowledge graph with entity extraction and relationship mapping.
AI-generated summary based on this skill's SKILL.md
Install
Egonex-AI/Understand-Anything/understand-knowledge · repository language: TypeScript
git clone https://github.com/Egonex-AI/Understand-Anything
cp -r Understand-Anything/understand-anything-plugin/skills/understand-knowledge ~/.claude/skills/understand-knowledgeFrequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
What does understand-knowledge do with my LLM wiki knowledge base?
understand-knowledge analyzes your LLM wiki knowledge base by processing three layers—raw sources, wiki markdown, and schema—to extract entities, resolve wikilinks, and cluster topics into an interactive graph. It detects the Karpathy wiki pattern, dispatches LLM agents to find implicit relationships, and merges results with deterministic extraction to output a navigable knowledge graph dashboard.
Can understand-knowledge generate a knowledge graph from markdown files?
Yes. understand-knowledge generates knowledge graphs directly from markdown files by extracting entities and relationships, resolving wikilinks with topic categorization, and clustering related concepts. The resulting graph is interactive and queryable, transforming your unstructured markdown documentation into a structured, navigable format.
How does understand-knowledge extract entities and relationships from wiki?
understand-knowledge uses a hybrid approach: it performs deterministic extraction of explicit entities and wikilinks from markdown, then dispatches LLM agents to detect implicit relationships that aren't directly stated. These results are merged to create a comprehensive entity and relationship map with topic layering and clustering.
What is the Karpathy pattern that understand-knowledge recognizes?
understand-knowledge recognizes the Karpathy-pattern LLM wiki structure—a specific organization of knowledge bases across raw sources, wiki markdown, and schema layers. When detected, it applies specialized processing to extract and visualize the knowledge graph with proper entity clustering and wikilink resolution.
Does understand-knowledge support wikilink resolution and graph generation?
Yes. understand-knowledge parses wikilinks and generates a graph by resolving links to their target topics, categorizing them by layer, and clustering related entities. This wikilink resolution feeds directly into the interactive knowledge graph visualization, making your wiki structure fully navigable.
What output does understand-knowledge produce?
understand-knowledge outputs an interactive knowledge graph dashboard that visualizes wiki structure, topics, and entity clusters. The graph is queryable and navigable, allowing you to explore relationships, browse topics, and understand your knowledge base structure at a glance.
SKILL.md
rendered from the published skill — quoted content, verbatim
/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
- Determine the target directory:
- If the user provided a path argument, use that
- Otherwise, use the current working directory -
(truncated - see the full file via the links below)
Read as markdown · JSON record · Browse the source repository
File tree — 3 files
understand-anything-plugin/skills/understand-knowledge/SKILL.md
understand-anything-plugin/skills/understand-knowledge/merge-knowledge-graph.py
understand-anything-plugin/skills/understand-knowledge/parse-knowledge-base.py