ai-agent

llm-wiki-skill: Build Knowledge Base

Name: llm-wiki-skill: Build Knowledge Base
Author: sdyckjq-lab

Construct a personal knowledge base with llm-wiki-skill.

2,375 stars285 forksTypeScriptGuide quality 8/10Updated 7/27/2026100% free · open source

What it does

Construct a personal knowledge base using the llm-wiki-skill tool, based on the Karpathy llm-wiki methodology, to organize and manage your knowledge across multiple platforms.

When to use it

•You need a centralized repository to store and connect your notes, ideas, and research.
•You want to build a knowledge graph that allows you to navigate and visualize relationships between concepts.
•You're looking for a tool to support your personal knowledge management and lifelong learning.

Ready-to-paste prompt

Use the command `npx ts-node src/cli.ts add-note --title 'My First Note' --content 'This is my first note.'` to create a new note in your knowledge base

Heads up: Make sure you have Node.js (14 or higher) and TypeScript installed on your system before attempting to install and run llm-wiki-skill, as it's built using TypeScript and relies on Node.js for execution.

Saves to your device

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/llm-wiki-skill && curl -fsSL https://workflowstacks.com/api/skills/llm-wiki-skill/claude-skill -o ~/.claude/skills/llm-wiki-skill/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How llm-wiki-skill: Build Knowledge Base works

Codeflow

Free to inspect

Llm-wiki-skill: Build Knowledge Base is a very large TypeScript project (~79k lines across 373 code files, plus 246 test files). It is a full software project: use it through its install path rather than reading it end to end. Last commit a month ago, no license file, has a test suite.

Size

Very large codebase

~79k lines · 373 code files · days to read — use, don't read

Setup

Developer setup

A real software project. Use it via its install path; don't expect to read it all.

Runs on

Node.js

No API keys detected

TypeScript 72%JavaScript 12%Python 6%Shell 6%CSS 2%

Where to start reading

1
README.md
Start here — what it does and how to install it
2
AGENTS.md
The instructions the AI actually follows
3
package.json
Dependencies and the commands it exposes

What's in each folder

packages/Sub-packages / plugins201 files

docs/Documentation137 files

templates/Templates14 files

workbench/Folder487 files

scripts/Helper scripts27 files

deps/Folder26 files

designs/Folder6 files

platforms/Folder5 files

READMEHas testsDocumentedCI checksNo licenseUpdated 1 mo ago

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/llm-wiki-skill && curl -fsSL https://workflowstacks.com/api/skills/llm-wiki-skill/claude-skill -o ~/.claude/skills/llm-wiki-skill/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

Quick Actions

Details

Creator

sdyckjq-lab

Language

TypeScript