A Free Interactive Platform by @satyatechgeek

AI writes the code.
You review it.

Interactive challenges for catching the bugs that Copilot, Cursor, and Claude Code sneak into your pull requests.

easyHallucinated APIs

Interactive Sample

The phantom module

A junior developer asked a chatbot to write a script that generates a cryptographically secure random identifier. The AI generated the following code, claiming it uses Node.js standard libraries. What is the issue here?

generate-id.js

const crypto = require('crypto');

function generateUserSessionId() {
  // Generate a random ID using standard crypto module
  const rawToken = crypto.generateRandomToken(32);
  return rawToken.toString('hex');
}

console.log(generateUserSessionId());

What is the primary bug in the AI-generated code?

crypto.generateRandomToken is not a standard Node.js API method; it should be crypto.randomBytes.

rawToken.toString() does not accept 'hex' as an encoding format; it only supports 'utf8'.

The crypto module must be installed via npm first; it is not a built-in Node.js module.

The function generates synchronous blockages and should be wrapped in an async/await block.

Challenges

Learn Specific Code Smells

Our challenges are divided into seven category patterns commonly produced by generative AI models.

Hallucinated APIs

LLMs invent modules, library arguments, or library functions that look real but don't exist.

Logic Errors

Incorrect boundaries, index offsets, pagination calculations, or missing edge cases.

Security

OS command injection, SQL injection, path traversal, or leaked credentials directly written into code.

Race Conditions

Stale React state closures, asynchronous listener leaks, or parallel request conditions.

Performance

N+1 query problems, heavy calculations running on render cycles, or database connection exhaustion.

Wrong Patterns

Silent failure handlers, anti-patterns, typing escapes, or incorrect React prop drilling patterns.

Dependencies

SSR window clashes, library version mismatch configurations, or incorrect peer configurations.

AI writes the code. You review it.