Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/chi-writing-style
Use when drafting or revising the prose of an ACM CHI paper — contribution statements, structure for mixed reviewer audiences, participant voice, calibrated claims, accessible and bias-free writing, and length discipline in a venue that reviews words against contribution instead of pages.
Passed 20 of 22 test cases with the skill, and 8 of 22 without it.
and 18 more in the full report
14 fixed6 passed either way0 still failing2 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 14 test cases the skill fixed. Same model, same prompt, with and without the skill.
Our full manuscript draft currently sits at 9,450 words in main.tex. Evaluate our word count against venue standards and provide the audit verdict block.
| Checked | Without skill | With skill |
|---|---|---|
| Assigns a length verdict of justify or cut for a word count above 8000 words | Missed | Passed |
| Includes the [Word count] line in the audit summary | Missed | Passed |
The instructions your model reads when the skill loads.
A CHI paper is read by a jury that rarely shares one methodological culture: a quantitative experimenter, a design researcher, and an ethnographer may all review the same submission. The prose must therefore do explicit work that a single-culture venue lets authors skip — naming the contribution type, justifying the method choice to outsiders, and calibrating claims to human evidence. And because CHI reviews length against contribution ("papers should be commensurate with their contribution"; 5,000–8,000 words encouraged, per the CHI 2027 Papers page, read 2026-07-08), every paragraph is auditable: does this paragraph earn its words?
CHI convention puts an explicit contribution list at the end of the introduction. Write it first, keep it honest, and let it govern the paper:
This paper contributes:
(1) <empirical finding>: the first interview study (n=24) of <population>
doing <activity>, showing <the finding in one clause>;
(2) <artifact>: <system>, an open-source <what>, which embodies <the design claim>;
(3) <design implications>: five implications for <who>, grounded in (1) and
tested against (2).
Each bullet must name its type (finding, artifact, method, theory — the taxonomy in
chi-topic-selection), its evidence, and its beneficiary. A bullet you cannot phrase
this way is not yet a contribution; cutting it now is cheaper than a reviewer doing
it in November.
| Section | The question it must answer | Style failure that costs papers |
|---|---|---|
| Abstract | What was done, with whom, what was found | Motivation essay; findings withheld like a plot twist |
| Introduction | Why this, why now, contributions | Three pages of stage-setting before any claim |
| Related work | Where this sits in HCI's conversation | Annotated bibliography instead of positioning (chi-related-work) |
| Method | Enough to replicate and to trust | Narrative gloss; missing recruitment, analysis, ethics |
| Findings/Results | Evidence, organized by claim | Organized by chronology of what the team did |
| Discussion | So what — for design, theory, practice | Restating results with "interestingly" attached |
test cases passed, vs 8 / 22 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.