AI Skill MartBrowse skillsPromptsProgramsLearnWhat is a Skill?
 
← Back to list

Knowledge Base Graph Analyzer (understand-knowledge)

Scans a Karpathy-pattern LLM wiki (raw sources + wikilinked markdown + schema) and builds an interactive knowledge graph with entities, implicit relationships, and topic clusters.

Data & AnalyticsAdvanced★ 91,543⑂ 7,812AI score 7/10Last updated: Nov 7, 2026

What it does

  • Detects whether a directory is a Karpathy-pattern wiki: an index.md plus multiple .md files containing [[wikilinks]], optionally a raw/ folder and a schema file (CLAUDE.md / AGENTS.md).
  • Runs the bundled parse-knowledge-base.py to deterministically extract wikilinks, headings, frontmatter, and index.md categories into scan-manifest.json.
  • Dispatches article-analyzer subagents in batches of 10–15 articles (up to 3 concurrently) to infer entities, implicit cross-references, and claims that plain parsing can't find.
  • Merges everything with merge-knowledge-graph.py, deduplicates entities, validates edges, then writes .ua/knowledge-graph.json plus meta.json.
  • Auto-triggers /understand-dashboard to render a force-directed graph view with category layers and a guided tour.

Who it's for

  • People running a personal knowledge base on Karpathy's three-layer LLM wiki pattern (raw / wiki / schema).
  • Anyone who wants to see structure and broken links (unresolved wikilinks) inside a heavily interlinked note collection.
  • Researchers and knowledge managers turning paper notes into a navigable topic map.

Examples

  1. /understand-knowledge ~/notes/ai-wiki — reports "N articles, N sources, N topics, N wikilinks (N unresolved)" and lists detected categories.
  2. Run with no argument to analyze the current directory; if it isn't a Karpathy wiki, the skill explains what was expected.
  3. After the merge, the dashboard opens with layers from index.md categories and a tour that follows the index section ordering.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/Egonex-AI/Understand-Anything/HEAD/understand-anything-plugin/skills/understand-knowledge/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on AI Skill Mart.
Copy the understand-anything-plugin/skills/understand-knowledge folder from the GitHub repo Egonex-AI/Understand-Anything into my ~/.claude/skills/understand-knowledge/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/Egonex-AI/Understand-Anything.git && mkdir -p ~/.claude/skills && cp -r Understand-Anything/understand-anything-plugin/skills/understand-knowledge ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Clone the repository: git clone https://github.com/Egonex-AI/Understand-Anything.git
  2. Locate understand-anything-plugin/skills/understand-knowledge/ and confirm it contains SKILL.md plus parse-knowledge-base.py and merge-knowledge-graph.py.
  3. Copy the whole folder into your skills directory: mkdir -p ~/.claude/skills && cp -r Understand-Anything/understand-anything-plugin/skills/understand-knowledge ~/.claude/skills/
  4. Verify Python 3 is available with python3 --version (required by the bundled scripts).
  5. For the automatic visualization step, also copy the understand-dashboard skill from the same plugin.
  6. Restart Claude Code and run /understand-knowledge <path-to-wiki>.
  7. Output lands in .ua/knowledge-graph.json inside the target directory (or .understand-anything/ if that folder already exists).
View source on GitHub ↗License: MIT