Install any skill in seconds. Free to start, no credit card required.
Get Started Free →rome-os/babysit-pr
Watch a PR for review feedback and answer each new round with the respond-to-review skill, on a self-paced timer, until the PR is approved or a round raises no new points.
Passed 10 of 16 test cases with the skill, and 10 of 16 without it.
3 fixed11 passed either way5 still failing4 broke
Tokens, turns and time describe resource use, not a dollar price.
This benchmark measured an earlier version of the skill.
One of 3 test cases the skill fixed. Same model, same prompt, with and without the skill.
During a check on PR #501, reviewer approvals are present with no unaddressed points, CI checks are green, and the branch merges cleanly. What scheduling action should be taken?
| Checked | Without skill | With skill |
|---|---|---|
| Stops scheduling further monitoring ticks. | Missed | Passed |
| Reports the outcome to the user. | Passed | Passed |
The instructions your model reads when the skill loads.
Watch $ARGUMENTS — the PR for the current branch when the argument is empty — and keep it ready to merge: run respond-to-review on every new round of review feedback, fix CI failures the diff caused, and resolve merge conflicts with the base branch. Merging the PR itself stays with the user.
The environment decides the mechanism: a watcher that wakes the session when a review or check lands, or a poll every five minutes. Each tick, handle whatever is new, then end the turn with the next tick scheduled. Once the PR is approved or a round raises no new points, CI is green, and the branch merges cleanly, schedule nothing and report the outcome. When the PR is blocked by another PR, schedule nothing, report which PR blocks it, and tell the user to run /babysit-pr again once that PR merges. When something needs the user's decision, stop and ask.
test cases passed, vs 10 / 16 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.