tkersey

LLM Wiki

A pattern for building personal knowledge bases using LLMs.

This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.

The core idea

Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.

karpathy

LLM Wiki

A pattern for building personal knowledge bases using LLMs.

This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.

The core idea

Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.

tkersey

2023

December

• MetalFX

Boost your Metal app’s performance by upscaling lower-resolution content to save GPU time.

The MetalFX framework integrates with Metal to upscale a relatively low-resolution image to a higher output resolution in less time than it takes to render directly to the output resolution. >

cameroncooke

# One-shot Codex: cdx "do x y and z"
cdx() {
  codex -a on-request \
    --sandbox danger-full-access \
    exec \
    --skip-git-repo-check \
    "$*" 2>/dev/null
}

intellectronica

Upstash Redis Key-Value Store SKILL

Get upstash-redis-kv.zip

Use this skill to read from and write to Upstash's Redis-compatible key-value store via REST API. Supports the full range of Redis data structures: strings, hashes, lists, sets, and sorted sets—ideal for caching, counters, leaderboards, queues, and persistent storage.

Works with Claude (by extracting it to ~/.claude/skills/) or with any other agent using Skillz.

Setup

intellectronica

YouTube Transcript SKILL

Get youtube-transcript.zip

Use this skill to fetch the transcript of a YouTube video, with or without timestamps.

Use this skill with Claude (by extracting it to .claude/skills/) or with any other agent using Skillz.

Note: This skill is unlikely to run successfully on the Claude web app, since access to YouTube is blocked. Use it with Claude Code or other local agents.

intellectronica

Nano Banana Pro SKILL

Use this skill to create or edit images using Nano Banana Pro.

Installation

Unzip nano-banana-pro.zip into ~/.claude/skills or ~/.skillz/ or wherever you keep skills locally. Or upload it to the Claude app.

intellectronica

GitHub Copilot CLI SKILL

Use models like Gemini 3 Pro, GPT-5.1, and GPT-5.1-Codex from within Claude by invoking GitHub Coplit CLI.

Installation

1. Create ~/.claude/skills/github-copilot
2. Save SKILL.md to ~/.claude/skills/github-copilot/SKILL.md

intellectronica

Markdown Converter Skill

Uses uvx markitdown to convert many document and file types to Markdown.

Great replacement for markitdown-mcp.

Works in Claude (upload markdown-converter.skill), Claude Code (expand the skill zip file into ~/.claude/skills/ or just create a skill directory with SKILL.md) or in any agent using Skillz.

steipete

• Delete unused or obsolete files when your changes make them irrelevant (refactors, feature removals, etc.), and revert files only when the change is yours or explicitly requested. If a git operation leaves you unsure about other agents' in-flight work, stop and coordinate instead of deleting.
• Before attempting to delete a file to resolve a local type/lint failure, stop and ask the user. Other agents are often editing adjacent files; deleting their work to silence an error is never acceptable without explicit approval.
• NEVER edit .env or any environment variable files—only the user may change them.
• Coordinate with other agents before removing their in-progress edits—don't revert or delete work you didn't author unless everyone agrees.
• Moving/renaming and restoring files is allowed.
• ABSOLUTELY NEVER run destructive git operations (e.g., git reset --hard, rm, git checkout/git restore to an older commit) unless the user gives an explicit, written instruction in this conversation. Treat t