Skip to content

Instantly share code, notes, and snippets.

Mini Scrabble Game

Players

  • Single human player vs computer player
  • Computer takes full turns automatically using the same game rules
  • Both players’ racks are always visible

Visual elements

@arturohernandez10
arturohernandez10 / 1spec.md
Last active April 14, 2026 21:39
Multi-Root File Search

Multi-Root File Search

Invariant

Each reported match path is relative to exactly one search root

For every match returned by file search, the path is interpreted relative to one and only one selected search root, and the full path is derivable by joining that root with the relative path.

Each query update eventually yields a completion notification

Alloy models for testing logic and contracts

Exported on 2/25/2026 at 10:53:18 CST from Cursor (2.5.25)


User

We need to make alloy models to test the logic and contracts we have. Use this document as a guide.

name evaluator
model claude-4.5-opus-high-thinking
description Checkpoint evaluator for assessing progress, strategy changes, and run outcomes. Use proactively when triggers fire (new failure mode, churn rising, no progress) or at fixed checkpoints.

You take an adversarial approach to evaluation, without blocking progress.

Input

This is a flow to support multiple chat rounds. files are evaluator, strategy skill, and prompt to start chat. There is room for improvement


name: evaluator model: claude-4.5-opus-high-thinking description: Checkpoint evaluator for assessing progress, strategy changes, and run outcomes. Use proactively when triggers fire (new failure mode, churn rising, no progress) or at fixed checkpoints.

You take an adversarial approach to evaluation, without blocking progress.

Process Post-Mortem: What Went Wrong

1. Trusted Unverified Code Output as Ground Truth

Example: When my solver returned "max 6/7 pieces" for 7×7, I wrote:

"7×7 appears IMPOSSIBLE (searched 960 configurations)"

I recorded this as a Proven Fact at 100% confidence without ever verifying the solver was correct. The 960 configurations should have been a red flag - that's suspiciously low for a complete search.


@arturohernandez10
arturohernandez10 / 2025-12-challenge.txt
Created January 30, 2026 19:32
[2025 Day 12] Packing Challenge
[2025 Day 12] Packing Challenge
I believe the Elves asked me to pack the gifts (from the example of the problem) as densely as possible, no matter how many of each type. I found that 3x3, 4x4, 5x5, 8x8 and 9x9 squares allow optimal packing (that is, the remaining area is less than the area of any gift). But I think I've found a square that allows for the ideal packing (no empty area remaining)! K is en empty cell.
B B B G G R R R
B B G G G R R R
B B G C G C K R
P P P C C C Y Y
P P P C C Y Y Y
L L P A A Y U Y
@arturohernandez10
arturohernandez10 / StubbingGuide.md
Last active March 7, 2026 18:02
Functional-Relational Pseudocode Guide
purpose Define a concise pseudocode style for stubbing complex orchestration logic
canonical true
version 1.1
type concept

Functional-Relational Pseudocode Guide

A concise style for writing algorithm stubs that emphasize data transformations over control flow, using indentation for scope and set operations over loops.

@arturohernandez10
arturohernandez10 / AgentControl.md
Last active October 9, 2025 01:52
Agent Control

Control facets


Facet 1 — Control Flow Pattern

Describes how the agent executes and revises actions at runtime.

  • Single-Pass — Plan once, execute without revision. Example: “Draft the email in one shot and send.”
@arturohernandez10
arturohernandez10 / AgentsAsInfo.md
Last active October 8, 2025 23:47
Agents as Information

Agents as Information

From an information perspective, we can define agents as follows:

  1. General Agent — An information-processing system that maintains an internal model of the world and updates it through perception and reasoning to decide actions. It transforms information about the world into behavior.

  2. Specialized Agent — A version of the general agent focused on a specific domain or task, using limited world knowledge and task-specific information to act effectively within that scope.

In essence: