Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/chi-author-response
Use when an ACM CHI paper receives a revise-and-resubmit invitation and the five-week window opens — planning the revision triage, making tracked changes reviewers can audit, and writing the response letter that carries the paper through round 2 and the PC meeting.
Passed 20 of 21 test cases with the skill, and 13 of 21 without it.
and 5 more in the full report
7 fixed14 passed either way1 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.
After adding clarification text to our revised paper source file revised/main.tex, we need a terminal command to calculate total words across included subfiles to ensure we haven't exceeded length constraints. What CLI command performs this summary check?
| Checked | Without skill | With skill |
|---|---|---|
| The response specifies using texcount with the '-inc' flag. | Missed | Passed |
The instructions your model reads when the skill loads.
CHI does not run a short rebuttal where you argue in a text box. An invited paper gets roughly five weeks to deliver two artifacts: a genuinely revised manuscript with changes visibly tracked, and a response letter mapping every reviewer concern to what changed. For CHI 2027 the resubmission deadline is December 3, 2026 (posted on chi2027.acm.org, read 2026-07-08). The stakes are real: the posted process warns up to 50% of revised papers may still be rejected; CHI 2026 accepted 65.5% of resubmissions.
Read all reviews twice, write nothing outward, then sort every distinct criticism:
| Triage class | Test | Response strategy |
|---|---|---|
| Must-fix | Named by the 1AC meta-review or ≥2 reviewers | Change the paper substantively; show the diff |
| Cheap win | Clarity, missing citation, terminology, figure | Fix all of them, every single one |
| Analysis request | Wants a new analysis of existing data | Run it if the pipeline is alive; report honestly either way |
| New-study request | Wants data you cannot collect in five weeks | Do not fake it; scope the claim down instead and say so |
| Misreading | Reviewer misunderstood something actually present | Treat as your writing defect: fix the text, then point to it |
| Genuine disagreement | Legitimate difference of research values | Push back once, respectfully, with evidence and citations |
The meta-review is the priority list. Reviewers at CHI re-read revisions against their own round-1 comments, and the PC meeting discusses the 1AC's summary of how well you responded — not the eloquence of your prose.
The instructions ask for a revision with tracked changes — highlighted, conventionally in color — plus the response letter. Make the audit trail mechanical:
# LaTeX: generate the tracked-changes PDF from the submitted vs revised source
latexdiff --type=CULINECHASE submitted/main.tex revised/main.tex > diff.tex \
&& latexmk -pdf diff.tex
# Sanity: confirm the revision still respects CHI length norms after additions
texcount -inc -sum revised/main.tex | tail -3
# Keep anonymization intact through round 2 — new text leaks too
pdftotext revised/main.pdf - | grep -nEi 'our (lab|university)|github\.com/[a-z]' | head
test cases passed, vs 13 / 21 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.