Skip to content

Instantly share code, notes, and snippets.

View mikiwiik's full-sized avatar

Miki Wiik mikiwiik

View GitHub Profile

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.

@MikaelSvenn
MikaelSvenn / gist:5fc7e40793f06b568373edafe89734ab
Created January 27, 2017 09:47
Guide for creating self-signed development certs using OpenSSL
Root-CA:
1) openssl genrsa -aes256 -out localhost-root-ca.key 4096
2) openssl req -out localhost-root-ca.csr -key localhost-root-ca.key -new -sha256
CN = localhost
-or-
CN = fully qualified domain name
3) openssl x509 -req -days 3000 -in localhost-root-ca.csr -signkey localhost-root-ca.key -out localhost-root-ca.pem -sha256
@davidlumley
davidlumley / client.rb
Last active March 23, 2020 18:42
Collection of XML requests and responses from AutoTasks SOAP API
=begin
======
example Savon client for connecting to AutoTask's SOAP API
- install savon via `gem install savon`
- ensure AUTOTASK_USERNAME and AUTOTASK_PASSWORD environment variables are set
=====
=end