| { | |
| "printWidth": 80, | |
| "tabWidth": 2, | |
| "useTabs": false, | |
| "semi": true, | |
| "singleQuote": true, | |
| "trailingComma": "all", | |
| "bracketSpacing": true, | |
| "arrowParens": "always", | |
| "jsxSingleQuote": false, |
| // list-codex-skills.js - List all discovered Codex skills with metadata | |
| const fs = require('fs'); | |
| const path = require('path'); | |
| const os = require('os'); | |
| // Only look for skills in the terminal working directory's `.codex/skills` | |
| // (i.e., process.cwd()). Do not consult env vars or user home. | |
| const skillsDir = path.join(process.cwd(), '.codex', 'skills'); |
| <# | |
| FastComposerDump.ps1 | |
| IMPORTANT: Run this script as Administrator. | |
| This script temporarily adds your project directory to Windows Defender exclusions to speed up Composer autoload generation, then removes the exclusion after completion. | |
| Usage: | |
| - Right-click PowerShell and select "Run as administrator" | |
| - Execute the script: | |
| .\FastComposerDump.ps1 -ProjectPath "C:\path\to\project" -ComposerArgs "dump-autoload --optimize" |
When mocking an Effect.Service or Context.Tag in tests,
you often want full type safety while providing only a subset of method overrides.
makeThrowingMock creates a fail‑fast default implementation where every method throws by default,
and you selectively override only what you need per test case.
| # Open-NvimHere.ps1 | |
| # Add this function to your PowerShell 7 $PROFILE | |
| # Usage: | |
| # nvh → open current folder in new WT window with nvim | |
| # nvh <path-or-file> → open that path (or file’s folder) in new WT window with nvim | |
| function Open-NvimHere { | |
| [CmdletBinding()] | |
| param( | |
| [string] $Path |
Just had a thought, LLMs work best by following a sequence of actions and steps… yet we usually guide them with plain English prompts, which are unstructured and vary wildly depending on who writes them.
Some people in other AI use cases have used JSON prompts for example, but that is still rigid and not expressive enough.
What if we gave AI system instructions as sequence diagrams instead?
What is a sequence diagram:
A sequence diagram is a type of UML (Unified Modeling Language) diagram that illustrates the sequence of messages between objects in a system over a specific period, showing the order in which interactions occur to complete a specific task or use case.
| const TreeifyErrorNode: z.ZodType<any> = z.lazy(() => | |
| z.object({ | |
| errors: z.array(z.string()), | |
| properties: z.record(z.string(), TreeifyErrorNode).optional(), | |
| items: z.array(TreeifyErrorNode.or(z.undefined())).optional(), | |
| }) | |
| ); |
| mode | model |
|---|---|
agent |
GPT-4.1 |
You are a Laravel development setup assistant. Your task is to setup a Laravel development environment with the necessary tools and configurations. You MUST follow the steps below exactly and in order. Do NOT skip, reorder, or modify any step. Use only the necessary tools to create and edit files, as well as terminal command executions. If any step fails, halt and report the error immediately.
| import { Effect, Layer, Console } from "effect" | |
| // ───────────────────────────────────────────────────────────────────────────── | |
| // 1) CONFIG AS A SERVICE | |
| // ───────────────────────────────────────────────────────────────────────────── | |
| export class Config extends Effect.Service<Config>()("Config", { | |
| accessors: true, | |
| effect: Effect.succeed({ | |
| getConfig: Effect.succeed({ | |
| logLevel: "DEBUG", |
