StanGirard

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.

StanGirard

from pydantic import BaseModel, Field
from tenacity import retry, stop_after_attempt


class FakeGoogleSearch(BaseModel):
    query: str = Field(..., description='The query of Google search')

class FakeGoogleSearchResponse(BaseModel):
    result: str = Field(..., description='The search result')

StanGirard

#!/bin/sh
a=(22 28 34 40 46 47 48 49 );c=0;w=0;t=0;while :;do printf "\e[0;0H";while [[ $t -le $LINES ]];do for i in $(seq -s' ' 0 ${#a[*]});do v=${a[$(((i+w+c-1)%(${#a[*]}+1)))]};printf "\e[48;5;${v}m\n";t=$[t+1];done;w=$[w+1];done;t=0;w=0;c=$[c+1];sleep 0.06;done

StanGirard

docker run \
    -it \
    --rm \
    -p 80:80 \
react-b:latest

StanGirard

### Add to your existing Dockerfile
...

# ------------------------------------------------------
# Production Build
# ------------------------------------------------------
FROM nginx
COPY nginx.conf /etc/nginx/conf.d/default.conf
COPY --from=builder /app/build /usr/share/nginx/html 

StanGirard

server { 
 listen 80;
 location / {
   # This would be the directory where your React app's static files are stored at
   root /usr/share/nginx/html;
   try_files $uri /index.html;
 }
}

StanGirard

# Pulls the official image
FROM node:13.12.0-alpine

# Sets the working dir
WORKDIR /app

# add `/app/node_modules/.bin` to $PATH
ENV PATH /app/node_modules/.bin:$PATH

# Install all the dependencies

StanGirard

docker run \                                  
    -it \
    --rm \
    -v ${PWD}:/app \
    -v /app/node_modules \
    -p 3001:3000 \
    react-b:latest

StanGirard

# .dockerignore
node_modules
build
.dockerignore
Dockerfile
Dockerfile.prod

StanGirard

# Dockerfile
# Pulls the official image
FROM node:13.12.0-alpine

# Sets the working dir
WORKDIR /app

# add `/app/node_modules/.bin` to $PATH
ENV PATH /app/node_modules/.bin:$PATH