Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when you need to address review or issue comments on an open GitHub Pull Request using the gh CLI.
.claude/skills/davila7-address-github-comments/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -40% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -62% | 0% |
| case-15 | ✓→✗ | ▼ Worse | -31% | 0% |
| case-13 | ✓→✓ | ＝ Same ✓ | -48% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | -46% | 0% |
Efficiently address PR review comments or issue feedback using the GitHub CLI (gh). This skill ensures all feedback is addressed systematically.
Ensure gh is authenticated.
bashgh auth status
If not logged in, run gh auth login.
Fetch the comments for the current branch's PR.
bashgh pr view --comments
Or use a custom script if available to list threads.
Apply the code changes for the selected comments.
Once fixed, respond to the threads as resolved.
bashgh pr comment <PR_NUMBER> --body "Addressed in latest commit."
gh auth status before starting.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-13 | pass→pass | 6,184 | 2,558 | -59% | 1 | 1 | 0% | 1,120 | 586 | -48% | 0 | 0 | — |
case-01 | fail→fail | 8,191 | 2,830 | -65% | 1 | 1 | 0% | 1,162 | 467 | -60% | 0 | 0 | — |
case-02 | fail→fail | 12,626 | 2,657 | -79% | 1 | 1 | 0% | 1,959 | 473 | -76% | 0 | 0 | — |
case-03 | fail→fail | 7,347 | 2,471 | -66% | 1 | 1 | 0% | 1,139 | 436 | -62% | 0 | 0 | — |
case-04 | fail→pass | 9,575 | 4,167 | -56% | 1 | 1 | 0% | 1,490 | 900 | -40% | 0 | 0 | — |
case-05 | pass→pass | 7,537 | 2,672 | -65% | 1 | 1 | 0% | 1,184 | 638 | -46% | 0 | 0 | — |
case-06 | pass→pass | 2,426 | 1,519 | -37% | 1 | 1 | 0% | 283 | 454 | +60% | 0 | 0 | — |
case-07 | pass→pass | 4,280 | 2,352 | -45% | 1 | 1 | 0% | 711 | 518 | -27% | 0 | 0 | — |
case-08 | pass→pass | 7,769 | 3,757 | -52% | 1 | 1 | 0% | 1,222 | 850 | -30% | 0 | 0 | — |
case-09 | pass→pass | 2,426 | 2,362 | -3% | 1 | 1 | 0% | 323 | 486 | +50% | 0 | 0 | — |
case-10 | pass→pass | 3,790 | 1,432 | -62% | 1 | 1 | 0% | 518 | 401 | -23% | 0 | 0 | — |
case-11 | pass→pass | 9,645 | 5,738 | -41% | 1 | 1 | 0% | 1,394 | 1,067 | -23% | 0 | 0 | — |
case-12 | fail→fail | 5,043 | 3,150 | -38% | 1 | 1 | 0% | 693 | 737 | +6% | 0 | 0 | — |
case-14 | pass→pass | 4,313 | 2,857 | -34% | 1 | 1 | 0% | 741 | 568 | -23% | 0 | 0 | — |
case-15 | pass→fail | 11,558 | 6,327 | -45% | 1 | 1 | 0% | 2,032 | 1,401 | -31% | 0 | 0 | — |
case-16 | pass→pass | 2,085 | 1,341 | -36% | 1 | 1 | 0% | 265 | 424 | +60% | 0 | 0 | — |
case-17 | pass→pass | 10,266 | 4,133 | -60% | 1 | 1 | 0% | 1,650 | 739 | -55% | 0 | 0 | — |
case-18 | fail→pass | 11,253 | 2,251 | -80% | 1 | 1 | 0% | 1,216 | 461 | -62% | 0 | 0 | — |
case-19 | pass→pass | 3,846 | 1,881 | -51% | 1 | 1 | 0% | 687 | 539 | -22% | 0 | 0 | — |
case-20 | pass→pass | 7,395 | 2,868 | -61% | 1 | 1 | 0% | 1,218 | 677 | -44% | 0 | 0 | — |
case-21 | pass→pass | 4,295 | 3,039 | -29% | 1 | 1 | 0% | 733 | 752 | +3% | 0 | 0 | — |
case-22 | pass→pass | 4,873 | 1,759 | -64% | 1 | 1 | 0% | 790 | 520 | -34% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of 0 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.