Use when organizing appendices and supplementary material for an ACL paper under ACL Rolling Review, covering the mandatory Limitations and optional ethics sections, appendices after references, anonymized software and data archives, the no-cloud-links rule, and deciding what must stay in the 8-page or 4-page body.
SkillScore86Maintained
Automatic
◆Research pack
Last verified 8/15/2026
What it changes
Passed 19 of 22 test cases with the skill, and 15 of 22 without it.
Without skill68%
With skill86%
Checks it passed that the model alone missed
Classifies the paper's split status as sound.
Notes that padding an unnecessary ethics statement invites unwarranted reviewer scrutiny.
Notes that omitting the Limitations section causes an automatic desk rejection under ACL rules.
and 3 more in the full report
4 fixed15 passed either way3 still failing0 broke
Pass-rate lift
+18 pp
Tokens
+16%
Agent turns
0%
Response time
-36%
Model gemini-3.6-flashMeasured 8/15/2026v119 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 4 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 4
Prompt
Our ACL paper introduces a standard Transformer optimization method tested on public synthetic benchmark matrices. We do not use human subjects, scraped user data, or demographic traits. Should we add a detailed Ethics Statement section to our paper? Assess our manuscript needs.
What was checked, without and with the skill
Checked
Without skill
With skill
Recommends omitting a dedicated Ethics Statement when no human subjects or realistic misuse risks are present.
Missed
Passed
Notes that padding an unnecessary ethics statement invites unwarranted reviewer scrutiny.
Missed
Passed
case-16 · gemini-3.6-flash · 8/15/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md117 linessingle file
ACL Supplementary
Use this when splitting an ACL paper between body, appendix, and archive. The
governing ARR principle: reviewers are not required to consider material in
appendices or supplements, so anything decision-critical that lives only
there is effectively invisible.
The ACL page anatomy
[ content pages: 8 long / 4 short ] <- the reviewed argument lives here
[ Limitations (REQUIRED, unlimited) ] <- after conclusion, outside page count
[ Ethics statement (optional) ]
[ References (unlimited) ]
[ Appendices (unlimited, same PDF) ] <- optional reading for reviewers
+ separate .tgz/.zip archive <- software / data supplement
Missing Limitations is a desk-reject condition; treating it as one throwaway
sentence is a review-stage penalty even when it passes the gate.
What must not leave the body
The main results table and the headline comparison.
Task definition and enough of the method that a reviewer can judge novelty.
At least a summary of the error analysis — a pointer-only error analysis
reads as not having one.
Human-evaluation design in one paragraph: raters, items, agreement.
The experimental setup at reproduction-outline level; full grids can go down.
What appendices are good at
Full prompt texts and few-shot exemplars (reference them per experiment).
Complete hyperparameter tables and search ranges.
Per-language / per-dataset breakdowns behind an averaged headline number.
Annotation guidelines and interface screenshots.
Extended qualitative examples and additional ablations.
Proofs or derivations for the occasional formal result.