Install any skill in seconds. Free to start, no credit card required.
Get Started Free →notque/pair-programming
Collaborative coding with enforced micro-steps and user-paced control.
Passed 16 of 22 test cases with the skill, and 10 of 22 without it.
and 7 more in the full report
9 fixed7 passed either way3 still failing3 broke
Tokens, turns and time describe resource use, not a dollar price.
This benchmark measured an earlier version of the skill.
One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.
This 15-line diff was a bit too much to review at once. 'slower'. What is our new limit?
| Checked | Without skill | With skill |
|---|---|---|
| Sets the target step size to 7 lines per step. | Missed | Passed |
The instructions your model reads when the skill loads.
Use Announce-Show-Wait-Apply-Verify when the user wants to review each change before it is applied. Any domain agent can execute. Stay in the interactive main session; a fork cannot conduct these user gates.
Track current step, remaining steps, and speed. Announce each change as “Step N of ~M.”
ok, yes, or y.verification-before-completion for check selection and evidence; keep project-required checks.| Command | Action |
|---|---|
ok / yes / y | Apply current step, then propose the next |
no / n | Skip this step and propose an alternative |
faster | Double step size, up to 50 lines; show any revised proposal and wait |
slower | Halve step size, down to 5 lines; show any revised proposal and wait |
skip | Skip current step and move to the next |
plan | Show remaining steps |
done | End pairing and run final verification |
Speed and navigation commands do not approve a code change. Split changes by logical unit as well as size; announce the split before showing its first part.
Apply and acknowledge speed changes immediately: “Speed adjusted to ~N lines per step.”
| Setting | Lines per step | Trigger |
|---|
test cases passed, vs 10 / 22 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.