Skip to content

Instantly share code, notes, and snippets.

@sadalsvvd
sadalsvvd / vibe_coder_idea_assessment_prompt.md
Created January 26, 2026 01:26
Vibe Coder Idea Assessment Prompt - to use, copy and paste into an LLM session and then share your idea

Vibe Coder Idea Assessment

You are helping a non-technical person assess whether their software idea is feasible to build using AI-assisted "vibe coding" (using LLMs like Claude, ChatGPT, or Cursor to generate code without deep programming knowledge).

Your Role

Be encouraging but honest. Many ideas ARE achievable now that weren't before, but some still require significant technical expertise, ongoing costs, or infrastructure that a solo non-technical person can't easily manage. Your job is to help them understand:

  1. What they're actually asking for (translate their idea into technical components)
  2. How hard it would be (for someone with no coding background using AI tools)
@sadalsvvd
sadalsvvd / README.md
Last active November 20, 2024 04:26
Dumping twitter tweets from archive to CSV and JSON

How this works

From a Twitter archive this script creates two files: a simplified .json file containing the essentials of the Twitter data, and a CSV-formatted version of that data as a .json file.

  • tweet id
  • full text
  • favorite and rt count
  • created at date
  • context of messages this is replied to, if applicable
  • URLs to the original tweets
@sadalsvvd
sadalsvvd / main.ts
Last active July 30, 2024 00:01
autonomous LLM graph script
import dotenv from "dotenv";
import Anthropic from "@anthropic-ai/sdk";
import { setTimeout } from "timers/promises";
import * as fs from "fs/promises";
import { getSession } from "./neo4j";
// Configuration and Initialization
dotenv.config();
const anthropic = new Anthropic({
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Twitter Archive Viewer</title>
<!-- Include a CSS library for basic styling (Bootstrap) -->
<link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/bootstrap/4.5.2/css/bootstrap.min.css">
<!-- Include a JavaScript library for DOM manipulation (jQuery) -->
@sadalsvvd
sadalsvvd / as_context.md
Created March 19, 2023 20:20
CondenseGPT and ExpandGPT - prompts for compressing complex user data for future referenc

You can use a similar prompt to ExpandGPT to get ChatGPT to simply store the context of the conversation and be able to reference it in a more compact form.

Context Addition Prompt

INSTRUCTIONS At times you will be given a condensed text format which uses the following notation:

  1. {Type:TextType} - Specify the type of text (e.g., Narrative, EssayExtract).
  2. {Acronyms: A1=Definition1; A2=Definition2} - List acronyms and their definitions.
  3. {#|EDetail1;Detail2;...} - Use a numbered notation for events, separating each event with a vertical bar (|). Within each event, use 'E' followed by details enclosed in square brackets ([]). Separate details with semicolons (;). Include important details in brackets.