Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/cscw-reproducibility
Use when strengthening the transparency of a CSCW paper — auditable qualitative analysis trails, documented trace pipelines, codebooks and instruments, and honest data-availability statements when community and participant data cannot ethically be shared.
Passed 22 of 23 test cases with the skill, and 10 of 23 without it.
and 12 more in the full report
12 fixed10 passed either way1 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 12 test cases the skill fixed. Same model, same prompt, with and without the skill.
In our social network trace dataset description, we listed the exact search queries and dates used for scraping. What additional context must be included in the Scope section of the trace ledger?
| Checked | Without skill | With skill |
|---|---|---|
| The scope section of the trace ledger requires explaining the rationale for choosing the specific query terms or time window. | Missed | Passed |
The instructions your model reads when the skill loads.
Reproducibility at CSCW cannot mean "rerun my script, get my table" — most of the venue's evidence is people, and much of it must never leave the research team. The venue's real standard is auditability: a skeptical reader should be able to see how you got from data to claims, and to build on the work, even where they cannot re-execute it. Different strands of a paper owe different transparency debts.
| Evidence strand | Shareable | Auditable instead of shareable |
|---|---|---|
| Interviews / fieldwork | Interview guide, recruitment text, codebook with definitions and example (paraphrased) excerpts | The analysis trail: coding approach, memo practice, how disagreements were resolved, how themes stabilized |
| Trace / log analysis | Pipeline code, query definitions, aggregated datasets, synthetic samples | Exact API/version/date of collection; filtering decisions with counts at each step; bot/deletion handling |
| Surveys | Full instrument, scale provenance, analysis scripts | Sampling frame, response/nonresponse accounting |
| Deployments | System code or architecture description, condition assignment logic | Site-selection reasoning; what the deployment context makes non-portable |
| Statistics anywhere | Analysis scripts keyed to each table/figure | Pre-specification vs. exploration, stated honestly |
You cannot share transcripts; you can share how you thought. The auditable minimum for interpretive work:
Platform data rots. Reviewers and future researchers need the ledger even when the data cannot travel:
test cases passed, vs 10 / 23 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.