Matt Silverlock elithrar

description: Planning agent that writes and updates plans to PLAN.md mode: primary permission: edit: PLAN.md: allow "": deny write: PLAN.md: allow "": deny

Code that uses await here for a synchronous call:

export class MyDurableObject extends DurableObject {
  async query() {
    const rows = await this.ctx.storage.sql.exec("select 1");
    return rows;
  }
}

agent-friendly-cli

Example CLI using @clack/prompts that outputs a machine-readable YAML schema when --help is invoked in a non-TTY context.

Context: https://x.com/elithrar/status/2008657921593536771

bun install
bun kv-cli.ts --help | cat  # YAML schema
bun kv-cli.ts # interactive mode

jj Cheat Sheet for Git Users

🗣️ this is a useful guide to help you memorize + better understand how Jujutsu works, but it's not a replacement for learning the mental model or deeper strategies!

Core Concepts

No staging area — all changes in the working copy are automatically part of the current “change”
Changes vs commits — jj tracks “changes” (mutable) that become “commits” (immutable) when pushed
@ — a symbol pointing to the current working copy change (like HEAD, but for changes)
@- — the parent of @

AGENTS.md

Instructions for AI coding agents working in this repository.

Decision-making

When possible, defer decisions to larger groups. Ensure all stakeholders have an opportunity to weigh in before proceeding.
Demand precision. Insist on clarifying ambiguities and edge cases before beginning any implementation.
Reopen previously settled matters if new participants join or if conditions have changed, even slightly.
Raise questions about whether the current approach aligns with long-term strategy. Suggest forming a working group if uncertainty remains.

	/**
	* Cross-Repo Tool for OpenCode
	*
	* Enables operations on GitHub repositories other than the current working repository.
	* Useful for coordinated changes across multiple repos, updating related repos,
	* or opening PRs in different repositories.
	*
	* ## Security Model
	*
	* Authentication is context-aware:

	import { Agent, AgentNamespace, routeAgentRequest } from 'agents';
	import { GoogleGenerativeAI, HarmCategory, HarmBlockThreshold } from '@google/generative-ai';

	// Define the environment variables and bindings required by the Worker and Agent
	interface Env {
	// Binding for the Agent Durable Object
	SECURITY_AUDIT_AGENT: AgentNamespace<SecurityAuditAgent>;
	// Secret holding the Google Gemini API Key
	GEMINI_API_KEY: string;
	// Optional: Secret holding a GitHub API token for higher rate limits / private repos

	<!DOCTYPE html>
	<html lang="en">
	<head>
	<meta charset="UTF-8">
	<meta name="viewport" content="width=device-width, initial-scale=1.0">
	<title>Simple Cloudflare Agent Chat</title>
	<style>
	body {
	font-family: Arial, sans-serif;
	max-width: 800px;

	{
	"$schema": "https://zed.dev/schema/themes/v0.1.0.json",
	"name": "Hypersubatomic",
	"author": "Converted from VSCode theme",
	"themes": [
	{
	"name": "Hypersubatomic Dark",
	"appearance": "dark",
	"style": {
	"background": "#0F111A",

	import { Solicitud, Respuesta, ContextoDeEjecucion, Entorno, Prometo, URL } from "./workers.es";

	export default {
	async obtener(solicitud: Solicitud, entorno: Entorno, contexto: ContextoDeEjecucion): Prometo<Respuesta> {
	let url = new URL(solicitud.url);
	let nombre = url.parametrosBusqueda.get("nombre") \|\| "mundo";

	return Respuesta.json({
	mensaje: `hola ${nombre}`,
	});