Tools

The test log

Every tool I’ve tested. Including the ones that flopped.

Each one got a real job and the same four questions. This page grows every time a new test video goes up, so it’s worth a bookmark.

1Tested
1Useful
0Not useful
USEFUL ✓ I’d keep using it, or pay for it. NOT USEFUL ✗ Overhyped, unreliable, or a free tool does it better.

Showing 1 tool

USEFUL ✓
Test no. 001 · Coding

Claude Code

An AI agent that works in your terminal. You describe what you want in plain English, and it reads your project, edits the files, runs the tests, fixes what breaks, and can open the pull request.

“Hours of coding work, done in minutes. As long as you read what it changed.”

Best for
Developers, and beginners building their first app
Price
Comes with paid Claude plans
Skip it if
You won’t review code before you ship it
How the test went
  1. The job: “Add dark mode, and fix the login bug.”
  2. Read the whole project before touching anything.
  3. Planned the changes and edited the files.
  4. Ran the tests. One failed.
  5. Read the error and fixed it without being asked.
  6. All tests passed. Committed and opened the pull request.
  7. Roughly 4 hours of work by hand, done in about 8 minutes.
Still on the list

Which one should I test next?

These parts of the log are empty. Reader suggestions go to the top of the pile.

Some links on this page are affiliate links. If you sign up through one, I may earn a commission at no extra cost to you. It never changes a verdict: a tool that fails gets a ✗ whether it pays or not. Full disclosure.