Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →archive228/bisect-regression
Find the exact commit that introduced a bug.
Passed 17 of 20 test cases with the skill, and 16 of 20 without it.
2 fixed17 passed either way3 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.
After running a search for a performance regression in `image-proc`, git output identified commit `c4d3e21` as the first bad commit. The commit diff changes 3 lines in `src/resample.c`. Someone suggests inspecting `src/pipeline.c` because it calls `resample.c`. Where should the search for the exact root cause line be confined?
| Checked | Without skill | With skill |
|---|---|---|
| Confines the bug search to the diff lines of commit c4d3e21. | Passed | Passed |
| Rejects inspecting unchanged files outside the commit diff. | Missed | Passed |
The instructions your model reads when the skill loads.
git bisect start; git bisect bad <bad>; git bisect good <good>.git bisect good/bad.git bisect reset. Report: the commit, the line, the one-sentence cause.
Automate it: git bisect run ./repro.sh if you have a script that exits non-zero on the bug.test cases passed, vs 16 / 20 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.