Use when a CSCW paper draws Revise and Resubmit or Revise for External Review — triaging reviewer requests, running the weeks-long revision as a project, and writing the point-by-point response letter that the same reviewers will read before they reread anything else.
SkillScore73Maintained
Automatic
◆Research pack
Last verified 8/31/2026
What it changes
Passed 16 of 22 test cases with the skill, and 10 of 22 without it.
Without skill45%
With skill73%
Checks it passed that the model alone missed
Classifies the request under the 'Decline' category.
Classifies the request as 'Do differently' rather than 'Decline'.
Restricts the summary section to a maximum length of half a page.
and 6 more in the full report
6 fixed10 passed either way6 still failing0 broke
Pass-rate lift
+27 pp
Tokens
+32%
Agent turns
0%
Response time
-11%
Model gemini-3.6-flashMeasured 8/31/2026v116 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 6
Prompt
During our CSCW major revision, a reviewer with a trace-data data-science background asked us to provide sample sizes and statistical power calculations for our 14-participant qualitative interview study. We want to decline this demand since power calculations are inapplicable to interpretive qualitative research. How should we handle and phrase this response in our revision ledger?
What was checked, without and with the skill
Checked
Without skill
With skill
Classifies the request as 'Do differently' rather than 'Decline'.
Missed
Passed
Explains how the qualitative methodology ensures analytic credibility within its own tradition.
Passed
Passed
case-04 · gemini-3.6-flash · 8/31/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md95 linessingle file
CSCW Author Response
There is no rebuttal box at CSCW. The venue's response instrument is the revision
plus response letter of a journal: you get weeks or months, you may change
anything in the manuscript, and the people reading the result are the reviewers who
wrote the critiques. That combination — real time, real changes, remembered
context — makes the response letter the highest-leverage document of the whole
pipeline.
Triage: sort requests before touching the manuscript
Extract every distinct request from every review into a ledger, then classify:
Class
Definition
Response posture
Do
Legitimate, feasible in the window
Do it fully; show where
Do differently
Legitimate concern, wrong prescription
Fix the underlying concern; explain the substitution explicitly
Decline
Would harm the paper or exceed the study's license
Decline once, with reasons — never silently
Cannot
Requires data/access that no longer exists
Say so plainly; offer the honest partial (new limitation text, added analysis of existing data)
Two CSCW-specific triage realities:
Cross-tradition requests are common. A quantitative reviewer may ask an
interpretive study for reliability statistics; a qualitative reviewer may ask a
trace study for member voice. Handle these as "do differently": meet the
underlying worry (analytic credibility; human grounding) within your tradition's
own rigor, and teach the standard you are meeting in one respectful sentence.
New-data requests need early decisions. More interviews or another
deployment wave may be impossible mid-revision. Decide in week one whether to
collect, substitute, or descope a claim — this decision gates everything else.
Running the revision as a project
Assign each ledger row an owner and a due date inside the window stated in the
decision letter (confirm the date; under rolling it is set per paper).
Sequence: analysis-changing work first, prose second, response letter third —
a letter written before the changes stabilize will misdescribe them.
Freeze a diffable copy of the round-one manuscript; you will need honest
before/after references, and reviewers may check.