A self-hosted, compounding-memory AI assistant running on a Raspberry Pi.
NanoClaw is a personal AI assistant built on Anthropic's Claude that runs entirely on a Raspberry Pi. It connects to messaging channels (WhatsApp, Telegram, Slack, Discord), processes voice and images, schedules recurring tasks, and — unlike a standard chatbot — accumulates knowledge over time through a structured memory system.
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.
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.
This prompt is designed for GPT-5, Claude Opus 4.1, Cursor, Claude Code, and cursor-agent to perform a deep, structured analysis of any codebase without pasting the entire repo into chat.
It guides the LLM to:
codebase-analysis-docs/
A simple "1-click" javascript approach to downloading a scanned book from archive.org to read at your leisure on the device of your choosing w/out having to manually screenshot every pages of the book by hand. In short it's a glorified "Save Image As..." approach but consolidated down to "1 click". BTW there may be a much better option than this out there - I just built this as an autistic project to see if it would work.
By using this script you agree to delete all book files/images after your 1 hour or 14 days is up! I don't support using this script for any other use cases. After all, none of us have ever kept a library book past it's return date, right?
| <!DOCTYPE html> | |
| <html> | |
| <head> | |
| <title>Timeline | Group example, with an arrow</title> | |
| <style> | |
| body, | |
| html { | |
| font-family: arial, sans-serif; | |
| font-size: 11pt; |
Get started:
mkdir hello-next
cd hello-next
npm init -y
npm install --save react react-dom next
mkdir pages
| Alright, thank you. | |
| So, I’ve got about apparently I’ve got about five to six minutes to say the most useful things I can think of. | |
| I’m gonna do my best. | |
| It was suggested that I distill things down to 3 items. | |
| I think I’ll go with four. | |
| And I’ll try, I think, I think these are pretty important ones. | |
| Some of it may kinda sound like, well you’ve heard them before. | |
| But, you know, worth reemphasizing. | |
| I think the first is, you need to work, if you, depending on how well you want to do, |
| CREATE TABLE `makers` ( | |
| `id` int(10) unsigned NOT NULL, | |
| `name` varchar(255) NOT NULL, | |
| `description` varchar(255) NOT NULL, | |
| `created_at` datetime NOT NULL, | |
| `updated_at` datetime NOT NULL, | |
| PRIMARY KEY (`id`) | |
| ) ENGINE=InnoDB DEFAULT CHARSET=latin1; | |
| -- |