Arthur CHAN arthurtalkgoal

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.

#How to Construct Yourself UI in KeystoneJS

KeystoneJS provide Admin UI with one set of route controllers and view templates(list&item) for all of the models.But usually,you will need some custome views other than Admin UI to display models. Although the KeystoneJS documention don't tell us much about how to contruct custome view,we can learn this from the source code in skeleton project generated by yo keystone,or you can just check out the keystone demo project's source code.We will walk through the blog feature's implementation in this demo application to demonstrate how to construct custome UI in KeystoneJS application.

As KeystoneJS documention described in Routes & Views section,there is a routes/index.js file, where we bind application's URL patterns to the controllers that load and process data, and render the appropriate template.You can find following code in it:

app.get('/blog/:catego

	# You need to install scikit-learn:
	# sudo pip install scikit-learn
	#
	# Dataset: Polarity dataset v2.0
	# http://www.cs.cornell.edu/people/pabo/movie-review-data/
	#
	# Full discussion:
	# https://marcobonzanini.wordpress.com/2015/01/19/sentiment-analysis-with-python-and-scikit-learn

	// Simulate config options from your production environment by
	// customising the .env file in your project's root folder.
	require('dotenv')().load();

	// Require keystone and i18n
	var keystone = require('keystone'),
	i18n= require('i18n');

	// Initialise Keystone with your project's configuration.
	// See http://keystonejs.com/guide/config for available options

	// Place this with the other middleware inclusion in routes/index.js

	keystone.pre('admin', middleware.enforcePermissions);