ndjordjevic

ndjordjevic / pin-llm-wiki

Public

Skill for Claude, Cursor & Copilot that automates the Karpathy LLM Wiki workflow: ingest web, GitHub, and YouTube URLs into a well-structured, citable, cross-referenced knowledge base with automatic linting.

agent-skill claude-skill cursor-skill llm-wiki
18
2
100% credibility
Found Apr 30, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Go Template
AI Summary

A tool for AI assistants in coding environments that builds a local, linked wiki from websites, GitHub repos, and YouTube videos, complete with summaries, citations, and instructions for agents.

How It Works

1
💡 Discover the wiki tool

You learn about a simple way to turn websites, videos, and code projects into your own personal knowledge collection that AI helpers can easily use.

2
📁 Create your wiki space

Make a new folder on your computer where all your collected knowledge will be organized and stored.

3
🛠️ Bring in the helper tool

Add this special tool to your folder so your AI assistant knows how to build and manage your wiki.

4
🚀 Kick off your wiki

Ask your AI to set everything up, and it creates starter pages, a queue for new items, and guides for using it.

5
Gather your sources
⚡
Add and process right away

Your AI fetches the content immediately, summarizes it, and weaves it into your wiki with links and notes.

📝
Save for a batch

Jot down the addresses in a handy list to handle several together when you have time.

6
🔗 Build and tidy up

Tell your AI to process everything, connect the pieces with helpful links, and check for any issues.

🎉 Use your smart wiki

Browse your easy-to-navigate knowledge base anytime, and watch your AI give precise, referenced answers drawn from it.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is pin-llm-wiki?

pin-llm-wiki is a GitHub skill that automates building a local knowledge base from web pages, GitHub repos, and YouTube videos, following the Karpathy LLM Wiki pattern. Drop URLs via slash commands in Claude Code, GitHub Copilot, or Cursor, and it fetches content into raw captures, generates summarized wiki pages with citations and wikilinks, plus an AGENTS.md for AI instructions. Written in Go Template, it creates inbox queues, linting for validation, and agent-ready structures without manual summarization.

Why is it gaining traction?

It stands out in the GitHub skill directory and marketplace by automating the full workflow—ingesting, cross-referencing, and linting—into a git-reviewable repo, unlike manual RAG setups or brittle chat contexts. Developers hook into it fast with `npx skills add` and commands like `/pin-llm-wiki ingest` or `queue`, supporting detail levels, branches, and multi-product splits for sites like LangChain. As a skill for Claude Anthropic, Copilot, and Cursor, it turns agent chats into persistent, citable memory.

Who should use this?

AI workflow engineers maintaining project docs from external sources, backend devs onboarding agents to GitHub repos or YouTube talks, and teams using Claude Code or Copilot for domain-specific queries. Ideal for research-heavy roles queuing videos or web docs into a shared wiki, or Cursor users needing automatic linting before agent responses.

Verdict

Early alpha with 18 stars and 1.0% credibility score—docs are solid but expect manual tweaks and no heavy testing yet. Worth a spin for Claude or Copilot power users prototyping agent knowledge bases; skip if you need production polish.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.