Urgent.News

What's breaking now, across thousands of outlets.

More in Tech

Green Tests Can Still Be Dead Tests: Auditing Agent-Generated Suites

A developer recently asked a coding agent to write a test suite for a function that parses ISO dates. The agent returned fourteen test cases, every one green, and the developer merged them without a…

  • Agent-generated test suite passed, but failed to cover new timestamp formats
  • Mutation testing reveals ineffective test suite with below 50% mutation score
  • MonkeyCode provides free mutation testing tools for agent-generated test suites

10 Million Free Tokens: A Token-Budget Field Test on a Free Server

A teammate received a free AI coding allowance last month. He burned it in two days. Not on complex architecture. On repeated full-file rewrites. Each rewrite consumed thousands of tokens.

  • Ten million tokens allow diverse AI coding tasks, from generation to refactoring
  • Budgetharness.py logs requests for token cost tracking in CSV format
  • Free server handles short tasks well but struggles with long completions

More from Monday 24 August →