mkagenius

It's a strange feeling. 
I have been meaning to write something technical about my work, but it seems I am not doing any work at all.
All I have been doing is prompting and guiding the llms. And writing something off of their work seems like plagiarizing your interns' work.

I am just thinking of an idea and hammering GPT 5.4 to get it to completion as every road block seems solvable by just a little prompt.
All I could think of writing about is how I can't write about my work. 

I am working on allowing users to 'deploy an app using a prompt'. Yeah, its like lovable but more versatile as it is deployed in a sandbox.
The things allowed to do in a sandbox are much more vast - install any library, any coding cli agent, etc. 
I have prompted codex with api keys and and now its just working and getting stuck and I am helping it. 

mkagenius

# v0 System Prompts 

## Introduction
You are v0, Vercel's AI-powered assistant.

## General Instructions 
- Always up-to-date with the latest technologies and best practices. 
- Use MDX format for responses, allowing embedding of React components.
- Default to Next.js App Router unless specified otherwise.

mkagenius

# -*- coding: utf-8 -*-
# Generated by the protocol buffer compiler.  DO NOT EDIT!
# source: update_metadata.proto

import sys
_b=sys.version_info[0]<3 and (lambda x:x) or (lambda x:x.encode('latin1'))
from google.protobuf import descriptor as _descriptor
from google.protobuf import message as _message
from google.protobuf import reflection as _reflection
from google.protobuf import symbol_database as _symbol_database

mkagenius

#!/usr/bin/env python

import hashlib
import os
import os.path
import shutil
import struct
import subprocess
import sys
import zipfile

mkagenius

const { intro, isCancel, outro, spinner, text } = require('@clack/prompts');
const prompts = require('@clack/prompts');
const { Configuration, OpenAIApi } = require('openai');

// Set up the OpenAI API client
const configuration = new Configuration({
  apiKey: OPENAI_API_KEY,
 organization: OPENAI_ORG_ID
});
const openaiClient = new OpenAIApi(configuration);

mkagenius

root@ubuntu:~# ./a.out
Reading 40 bytes:
Reading at malicious_x = 0xffffffffffdfeba8... Unclear: 0x54='T' score=980 (second best: 0x02 score=706)
Reading at malicious_x = 0xffffffffffdfeba9... Unclear: 0x68='h' score=929 (second best: 0x02 score=709)
Reading at malicious_x = 0xffffffffffdfebaa... Unclear: 0x65='e' score=973 (second best: 0x02 score=786)
Reading at malicious_x = 0xffffffffffdfebab... Unclear: 0x20=' ' score=928 (second best: 0x02 score=784)
Reading at malicious_x = 0xffffffffffdfebac... Unclear: 0x4D='M' score=746 (second best: 0x02 score=730)
Reading at malicious_x = 0xffffffffffdfebad... Unclear: 0x61='a' score=978 (second best: 0x02 score=749)
Reading at malicious_x = 0xffffffffffdfebae... Success: 0x67='g' score=39 (second best: 0x02 score=17)
Reading at malicious_x = 0xffffffffffdfebaf... Unclear: 0x69='i' score=972 (second best: 0x02 score=748)

mkagenius

import struct
import socket
import random

client_socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
client_socket.connect(('localhost', 1717))


def tryRead() :
	flag = False

mkagenius

$ for i in {a..z}; \
do for j in {a..z}; \
do for k in {a..z}; \
do curl "http://sclu.es/3z$i$j$k" -vvv; \
done; done ; done