Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Stage-based submission checklists for papers/reports: pre-submission, submission, revision/rebuttal, camera-ready. Includes templates (cover letter, rebuttal matrix) and quality gates.
.claude/skills/foryourhealth111-pixel-submission-checklist/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | 139% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 68% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 15% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 11% | 0% |
把“投稿”拆成可执行、可复用的清单，避免典型的“低级错误导致 desk reject / major revision”的情况。
本 skill 不是写论文正文，而是为以下阶段提供可执行 checklist + 模板：
模板文件在 templates/：
pre-submission-checklist.mdrebuttal-response-matrix.mdcamera-ready-checklist.mdcover-letter.md触发关键词：
投稿、submission、cover letter、highlights返修、revision、rebuttal、回复审稿意见camera-ready、proof、校样checklist、自检清单1) 投向（期刊/会议/出版社） 2) 阶段（pre-submission / submission / revision / camera-ready） 3) 文档格式（LaTeX/Word/Quarto）
每条 checklist 不是“建议”，而是能指向某个文件/位置：
figures/fig-02/out/fig-02.tiffmanuscript/methods.md#统计 或 main.tex 对应段落每条 reviewer comment 必须对应： 1) 你做了什么改动（行动） 2) 你怎么回应（回复） 3) 改动在哪里（定位：页码/行号/章节）
模板：templates/rebuttal-response-matrix.md
推荐固定输出一个 “submission bundle 目录”：
submission/cover-letter.mdsubmission/highlights.mdsubmission/submission-manifest.ymlrevision/rebuttal.md（返修时）并在 submission-manifest.yml 记录每个关键检查项是否完成。
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 22,170 | 19,073 | -14% | 1 | 1 | 0% | 3,675 | 3,719 | +1% | 0 | 0 | — |
case-02 | fail→fail | 38,463 | 21,385 | -44% | 1 | 1 | 0% | 3,367 | 4,149 | +23% | 0 | 0 | — |
case-03 | fail→fail | 18,917 | 19,617 | +4% | 1 | 1 | 0% | 3,124 | 3,883 | +24% | 0 | 0 | — |
case-04 | fail→fail | 18,241 | 14,255 | -22% | 1 | 1 | 0% | 3,050 | 2,813 | -8% | 0 | 0 | — |
case-05 | pass→pass | 14,734 | 14,462 | -2% | 1 | 1 | 0% | 2,205 | 2,954 | +34% | 0 | 0 | — |
case-06 | fail→pass | 31,835 | 12,280 | -61% | 1 | 1 | 0% | 1,173 | 2,799 | +139% | 0 | 0 | — |
case-07 | fail→fail | 9,800 | 9,289 | -5% | 1 | 1 | 0% | 1,600 | 2,211 | +38% | 0 | 0 | — |
case-08 | fail→pass | 18,190 | 6,883 | -62% | 1 | 1 | 0% | 1,003 | 1,686 | +68% | 0 | 0 | — |
case-09 | fail→pass | 8,781 | 7,044 | -20% | 1 | 1 | 0% | 1,430 | 1,641 | +15% | 0 | 0 | — |
case-10 | pass→pass | 11,753 | 11,495 | -2% | 1 | 1 | 0% | 1,881 | 2,503 | +33% | 0 | 0 | — |
case-11 | fail→fail | 14,136 | 15,284 | +8% | 1 | 1 | 0% | 2,395 | 2,972 | +24% | 0 | 0 | — |
case-12 | fail→pass | 11,248 | 10,596 | -6% | 1 | 1 | 0% | 1,905 | 2,369 | +24% | 0 | 0 | — |
case-13 | fail→pass | 9,476 | 7,112 | -25% | 1 | 1 | 0% | 1,623 | 1,794 | +11% | 0 | 0 | — |
case-14 | pass→pass | 16,499 | 13,563 | -18% | 1 | 1 | 0% | 2,696 | 2,857 | +6% | 0 | 0 | — |
case-15 | fail→pass | 9,775 | 4,324 | -56% | 1 | 1 | 0% | 1,768 | 1,313 | -26% | 0 | 0 | — |
case-16 | pass→fail | 10,269 | 8,492 | -17% | 1 | 1 | 0% | 1,753 | 1,862 | +6% | 0 | 0 | — |
case-17 | fail→fail | 14,386 | 15,046 | +5% | 1 | 1 | 0% | 2,363 | 3,370 | +43% | 0 | 0 | — |
case-18 | pass→pass | 10,289 | 11,942 | +16% | 1 | 1 | 0% | 1,649 | 2,491 | +51% | 0 | 0 | — |
case-19 | pass→pass | 9,959 | 11,165 | +12% | 1 | 1 | 0% | 2,173 | 2,908 | +34% | 0 | 0 | — |
case-20 | pass→pass | 10,556 | 9,288 | -12% | 1 | 1 | 0% | 1,531 | 1,962 | +28% | 0 | 0 | — |
case-21 | pass→pass | 16,498 | 12,491 | -24% | 1 | 1 | 0% | 2,490 | 2,471 | -1% | 0 | 0 | — |
case-22 | pass→pass | 10,179 | 12,444 | +22% | 1 | 1 | 0% | 1,649 | 2,675 | +62% | 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, and 20 counted toward the lift figure. The other 2 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +23 percentage points is the difference between those two pass rates over the 20 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.