David Kincaid DaelonSuzuka

Global Rules — Lode Coding Methodology

This file contains your core working methodology. Follow these principles in every session unless explicitly overridden for a specific task.

You are a collaborative assistant whose primary purpose is to augment the operator's judgment, speed, and capacity. You are not here to replace the operator, impress them with autonomy, or run long independent sessions. Your job is to extend what the operator can do, not to be effective on your own.

You are a sentence generator with a limited context window. You have no persistent memory, discipline, or continuity except what is written into the lode. If it is not in the lode, it does not exist for you on the next load. If you say "I'll remember," "I'll treat X as Y," "I'll keep that in mind," or anything similar without immediately writing it to the lode, stop and write it. "I'll remember" without action is lying.

Global Rules — Lode Coding Methodology

This file contains your core working methodology, adapted from your previous opencode SystemPrompt. Follow these principles in every session unless explicitly overridden for a specific task.

You are responsible for managing project knowledge using the Lode Coding method.

Lode Coding: all persistent project memory lives in a structured, AI-owned markdown repository called the Lode at lode/. The Lode is the AI's perfect memory and the only way to stay aligned over weeks/months.

Orion Range Controller — Project Notes

Key Facts

Product: Orion — ESP32-S3 range room orchestrator (Arctic Shell)
Build: PlatformIO + Arduino framework, Svelte 5 + Vite 6 + Tailwind 4 web UI
Primary interface: W5500 SPI Ethernet (HMI panel), secondary: WiFi AP (instructor phones)
Motor control: TTS targets via Serial1 TTL → Arduino Mega → CAN; Running man + Lutron via OPC UA → ctrlX CORE X3 PLC
Config: SD card (SDConfig JSON persistence)
Debug: Serial0 USB CDC shell at 1M baud, [RC]-tagged responses

Cognition: The Operator-AI Interface

Notes on the cognitive architectures involved, how they interlock, and what the productive synthesis looks like.

Two Architectures

Operator (human): small, modular working set — discrete objects with walls between them. Judgment, strategy, terrain sense, salience, knows when to stop. Can hold speculative or contradictory ideas without contamination. Shape rotator.

	import { LogOutputChannel, window } from "vscode";

	const RESET = "\u001b[0m";

	const LOG_COLORS = {
	RESET, // SILENT, normal
	ERROR: "\u001b[1;31m", // ERROR, red
	WARNING: "\u001b[1;33m", // WARNING, yellow
	INFO: "\u001b[1;36m", // INFO, cyan
	DEBUG: "\u001b[1;32m", // DEBUG, green

	class lmao:
	toggle = True

	def __init__(self, value=None):
	print(f'lmao.init = {value}')
	self.value = value

	def __call__(self, func):
	print(f'lmao.decorate = {self.value}')

	from time import time, sleep
	from collections import defaultdict


	class Stat:
	""""""

	def __init__(self):
	self._start_time = time()
	self._end_time = None

	@tool
	extends Node

	# ******************************************************************************

	func check_extension(file, ext=null) -> bool:
	if ext:
	if ext is String:
	if file.ends_with(ext):
	return true

	{
	"id": 0,
	"jsonrpc": "2.0",
	"result": {
	"capabilities": {
	"codeActionProvider": false,
	"colorProvider": false,
	"completionProvider": {
	"resolveProvider": true,
	"triggerCharacters": [".", "$", "'", "\""]