webconf-reproducibility
Use when hardening the reproducibility of a Web Conference (WWW) paper whose evidence rests on crawls, platform APIs, live systems, or user logs — covering dataset decay, temporal snapshots, seed and environment reporting, the reproducibility appendix inside the 12-page PDF, and honest claims when the Web itself cannot be replayed.
Profile →npx skills add brycewang-stanford/Awesome-Journal-Skills --skill webconf-reproducibility --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# Web Conference Reproducibility Reproducibility at this venue has a problem no offline-ML venue has: **the object of study mutates**. Pages die, APIs close, ranking systems retrain, platform policies change what may be collected at all. A Web Conference paper is reproducible to the degree that it *pins what can be pinned and measures what cannot*. The 2026 CFP's sanctioned home for this material is the optional appendix — "details on reproducibility, proofs, pseudo-code" — inside the same 12-page PDF, which reviewers are not obliged to read; so the reproducibility *claims* go in the main 8 pages and the reproducibility *mechanics* go in the appendix. ## Three reproducibility regimes | Regime | Example evidence | What "reproducible" means | Your obligation | |---|---|---|---| | Frozen | Public benchmark, released crawl | Re-run → same numbers | Seeds, versions, exact splits | | Decaying | Your own crawl, API pulls | Re-collect → quantifiably similar corpus | Snapshot, checksums, collection code, date stamps | | Unreplayable | Live A/B test, production traffic, human subjects | Independent teams can audit the protocol | Full protocol, power analysis, aggregate release | Most reviews
- Three reproducibility regimes
- Pinning the decaying regime
- Determinism audit for the modeling stack
- Temporal honesty
- Vignette: the benchmark that dissolved
- What goes where
- Pre-submission reproducibility gate
- Output format
What does the webconf-reproducibility skill do?
Use when hardening the reproducibility of a Web Conference (WWW) paper whose evidence rests on crawls, platform APIs, live systems, or user logs — covering dataset decay, temporal snapshots, seed and environment reporting, the reproducibility appendix inside the 12-page PDF, and honest claims when the Web itself cannot be replayed.
How do I install it?
Run `npx skills add brycewang-stanford/Awesome-Journal-Skills --skill webconf-reproducibility --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From brycewang-stanford/Awesome-Journal-Skills, a repository with 909 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.