Test results

CodeTour File Builder: test results

Tested 2026-10-19, skill version 1.0.0 at the time of loading this page. We run every skill on a strong and a weak model before it is listed, and publish both verdicts, including where the weak one fails.

Verdicts

Date
2026-10-19
Strong · claude-sonnet-5-5 (Claude Code alias "sonnet")
Right on all 3 cases, read by hand. On a login walkthrough it kept three steps, left out the stop whose file is not in the list without guessing the similar logger file, and named it in the description in the fixed form. On a billing walkthrough it cleaned the leading dot and the backslashes, kept the steps in order and skipped only the line past the end. On a clean tour it kept the step on the last line, left out the line for the step that had none and added no description. Every answer was bare JSON.
Weak · claude-haiku-5-5 (Claude Code alias "haiku")
Right on all 3 cases, read by hand. Its three answers were identical in content to the strong model: the missing file named in the fixed form, the path spellings cleaned, the last line kept and no description on the clean tour. Bare JSON each time. The inputs resemble the example inside the skill, so this is a light confirmation, not a hard test.

Note

Light test: three cases (two planted, one clean), one run per model; no comparison without the skill. The three inputs were written by us; each planted one carries a safe look-alike and neither planted case has a comment pointing at the defect. Facts come from public documentation, read on 19 October 2026; there are no owner-measured facts.

What was not measured

  • Models other than the two named above were not run.
  • Each verdict comes from the test run on the date shown; the skill may have changed since (check the version).
  • Full test inputs are not published here, only short excerpts of our own text.
  • Results on your own texts, languages and domains can differ.

Back to CodeTour File Builder · Card (JSON)