New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Kdd Reproducibility

by brycewang-stanford in brycewang-stanford/Awesome-Journal-Skills

A skill your agent uses when hardening the reproducibility story of a KDD paper, where reproducibility is an explicit decision factor for area chairs.

MITAuto-check passedResearch & Science

Install Kdd Reproducibility

skills CLI
$ npx skills add brycewang-stanford/Awesome-Journal-Skills --skill kdd-reproducibility -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install brycewang-stanford/Awesome-Journal-Skills kdd-reproducibility --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/brycewang-stanford/Awesome-Journal-Skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/KDD-Skills/skills/kdd-reproducibility .claude/skills/kdd-reproducibility && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
kdd-reproducibility
GitHub stars
1.2k
Token cost
~1.8k tokens
SKILL.md length
780 words
Files
1
Skills in repo
2,387
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses when hardening the reproducibility story of a KDD paper, where reproducibility is an explicit decision factor for area chairs.

  • Works in 3 steps: Rerunnable — public data, shipped code,… → Rebuildable — method fully specified… → Attested — post-launch/production…
  • Hardening the reproducibility story of a KDD paper
  • SKILL.md covers Three-tier honesty model, Mining-pipeline disclosure…, Config-as-artifact discipline and Stochasticity on large data, plus 4 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Kdd Reproducibility is an agent skill from brycewang-stanford/Awesome-Journal-Skills. Use when hardening the reproducibility story of a KDD paper, where reproducibility is an explicit decision factor for area chairs. Covers claim-to-evidence tiers for mining pipelines, seeds and splits on large graphs and streams, baseline-tuning disclosure, compute reporting, and honest limits for ADS deployment results.

Its SKILL.md is about 1.8k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Research & Science, covering Reproducible research. The repository describes itself as: Journal-specific Claude Code/Codex skill packs covering mainstream journals — AER, QJE, Nature, Cell, 管理世界, 经济研究 & 200+ more — your fast track to getting published. ｜ 覆盖主流期刊的… The licence is MIT.

When your agent uses it

  • Hardening the reproducibility story of a KDD paper
  • Where reproducibility is an explicit decision factor for area chairs

Example prompts

  • “/kdd-reproducibility”

Requirements

  • Python 3

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. Rerunnable — public data, shipped code, fixed seeds; a reader regenerates the
  2. Rebuildable — method fully specified (pseudocode + hyperparameters + data
  3. Attested — post-launch/production measurements that no outsider can repeat;

What it can do on your machine

Read from SKILL.md and the folder at commit 932eb23. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are python).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Kdd Reproducibility loads about 1.8k tokens when it runs. Until then it costs about 86 tokens; SKILL.md has 780 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~86
When it runs · the whole SKILL.md, loaded when a task matches
~1.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from brycewang-stanford/Awesome-Journal-Skills at commit 932eb23, republished under its MIT licence (© brycewang-stanford). 780 words, ~1,771 tokens.

Download SKILL.mdSave it as .claude/skills/kdd-reproducibility/SKILL.md (or your agent's skills folder).
name
kdd-reproducibility
description
Use when hardening the reproducibility story of a KDD paper, where reproducibility is an explicit decision factor for area chairs. Covers claim-to-evidence tiers for mining pipelines, seeds and splits on large graphs and streams, baseline-tuning disclosure, compute reporting, and honest limits for ADS deployment results.

KDD Reproducibility

Use this before the paper freezes. The KDD 2026 CFP lists reproducibility of results among the factors area chairs weigh in acceptance recommendations — at this venue it is a scored dimension, not a virtue. The bar is shaped by what KDD papers claim: pipelines over large, messy, sometimes proprietary data, where the reader must be able to tell exactly which parts they can rerun, which they can rebuild, and which they must take on documented trust.

Three-tier honesty model

Declare every result in the paper as one of:

  1. Rerunnable — public data, shipped code, fixed seeds; a reader regenerates the number. This should cover the headline comparison table.
  2. Rebuildable — method fully specified (pseudocode + hyperparameters + data schema), but data is proprietary or too large; a reader can reconstruct the pipeline on their own data.
  3. Attested — post-launch/production measurements that no outsider can repeat; reproducibility here means the measurement protocol is fully specified.

An ADS paper is usually tier 2-3; a Research Track paper claiming tier 1 while shipping tier 2 is what burns trust in review.

Mining-pipeline disclosure checklist

The stages that silently decide results in data-mining papers, and what to pin:

Pipeline stageWhat must be pinnedClassic KDD reviewer catch
Data acquisitionVersion/date of each dataset, filtering rules, dedup"Which snapshot of the graph?"
PreprocessingFeature construction code, normalization, leakage guardsTarget leakage via time-travel features
SplitsSplit logic (temporal vs random), exact seeds, cold-start handlingRandom splits on temporal data inflating results
Negative samplingRatio, distribution, per-epoch resampling or notBaselines run with a different sampling scheme
BaselinesSource of implementation, tuning budget per baselineOwn method tuned for weeks, baselines run at defaults
Hardware/computeMachines, memory, wall-clock per experimentThroughput claims with no hardware context

Equal-tuning-budget disclosure matters more at KDD than almost anywhere: the venue's history is full of boosting/embedding/GNN comparisons decided by tuning asymmetry.

Config-as-artifact discipline

Every reported number should trace to a committed config, and the mapping should be mechanical:

python
# repro/manifest.py - emitted next to every result file
import json, subprocess, time

def write_manifest(cfg, dataset_stats, out="results/manifest.json"):
    manifest = {
        "config_file": cfg.path,              # one config per table row
        "seed": cfg.seed,                     # and the full seed list for repeats
        "dataset": dataset_stats,             # {"name":..., "rows":..., "edges":..., "snapshot":...}
        "git_commit": subprocess.check_output(
            ["git", "rev-parse", "HEAD"]).decode().strip(),
        "wall_clock_sec": cfg.elapsed,
        "hardware": cfg.hardware,             # e.g. "1x A100-80GB, 256GB RAM"
        "timestamp": time.strftime("%Y-%m-%dT%H:%M:%SZ"),
    }
    json.dump(manifest, open(out, "w"), indent=2)

Tables generated from manifests cannot drift from the artifact — the failure mode where the PDF says 0.847 and the repo reproduces 0.831 is a rebuttal-phase disaster, and KDD rebuttals cannot even link to a corrected artifact.

Stochasticity on large data

  • Repeat runs are expensive at KDD scale; the honest compromise is repeats on the small/medium datasets plus a variance statement, and single seeded runs at the largest scale with the seed published.
  • State what is nondeterministic even under a fixed seed (parallel reductions, GPU atomics, hash-partitioned sampling), so a near-miss reproduction reads as expected rather than as failure.
  • Never present a best-of-k run as a typical run; if model selection used validation metrics, say k and the selection rule.
Show full SKILL.md (332 more words)Show less

ADS attested-results protocol

For post-launch numbers, reproducibility means specification: metric definitions, measurement window, traffic allocation, ramp schedule, guardrails, and any seasonal confounders in the window. A reader should be able to audit the measurement, even though they cannot repeat it. Blur only what confidentiality forces (absolute denominators can become relative lifts), and say explicitly what was blurred and why.

Vignette: the tier audit that changed a claim

A recommendation paper reports wins on two public datasets and one "large industrial dataset." The tier audit finds: public results are rerunnable (good); the industrial result is attested but the paper's abstract says "we release all code and data" — written before the legal review pulled the industrial set. Left uncorrected, that sentence is the kind of paper-vs-reality contradiction a practitioner reviewer catches in minutes and generalizes from ("what else is oversold?"). The repair:

  • Abstract sentence rewritten to "code and the two public datasets are released; the industrial pipeline is fully specified in Section 4."
  • The industrial section gains the schema, size, and collection-window spec that makes it rebuildable-in-principle.
  • The headline claim re-anchored on the public results, with the industrial run as corroboration rather than as the primary evidence.

Reproducibility review at KDD is largely consistency review: the checklist is the paper against itself.

Pre-freeze reproducibility gate

Run this list the week before the paper freezes, while fixes are still cheap:

  1. Every number in the abstract traces to a table, and every table row traces to a committed config.
  2. Tier labels (rerunnable/rebuildable/attested) assigned per result and reflected in the availability sentence — no tier inflation anywhere in the text.
  3. Split logic re-derived from the code, not from memory; temporal data confirmed time-ordered end to end.
  4. Baseline provenance table complete: implementation source, version, tuning budget.
  5. The manifest of the single largest run (hardware, wall-clock, dataset cardinality) exists — it anchors both the scale claim and the compute disclosure.
  6. Repository smoke-tested from a clean clone by someone who did not write it.

Output format

text
[Tier map] <headline table: rerunnable/rebuildable/attested, per major result>
[Pipeline pins] <acquisition/preprocessing/splits/sampling/baselines/compute status>
[Tuning symmetry] <our budget vs baseline budget, disclosed where>
[Variance story] <repeats + seeds at which scales>
[ADS measurement spec] complete / gaps: <...> / N-A
[Trust risks] <paper-artifact drift, leakage, undeclared selection>

© brycewang-stanford, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in KDD-Skills/skills/kdd-reproducibility of brycewang-stanford/Awesome-Journal-Skills.

Open the folder on GitHubat commit 932eb23

Compare with similar skills

Kdd Reproducibility next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Kdd Reproducibility compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Kdd Reproducibility this skillbrycewang-stanford/Awesome-Journal-Skills1.2k—~1.8kAutomated safety check: PassMIT
Peer ReviewK-Dense-AI/claude-scientific-writer2.4k2 repos~3.1kAutomated safety check: NotesMIT
CHARLS Paper Reproduction Guidexjtulyc/MedgeClaw6171 repos~1.8kAutomated safety check: PassNone
Compute Environment Setupaipoch/open-science5.5k—~2.6kAutomated safety check: PassApache-2.0
Figure Styleaipoch/open-science5.5k—~5.1kAutomated safety check: PassApache-2.0
Add Bactopia Toolbactopia/bactopia522—~4.1kAutomated safety check: PassMIT

Similar skills

  • Peer Review

    K-Dense-AI/claude-scientific-writer

    Prepare evidence-bounded, constructive peer-review drafts and structured manuscript assessments.

    2.4k GitHub starsUsed in 2 repos~3.1k tokens
    Research & ScienceAuto-check: notes

  • CHARLS Paper Reproduction Guide

    xjtulyc/MedgeClaw

    Guides an agent through reproducing papers built on the CHARLS health and retirement survey, from variable mapping to cognition, depression and isolation scores.

    617 GitHub starsUsed in 1 repo~1.8k tokens
    Research & ScienceAuto-check passed

  • Compute Environment Setup

    aipoch/open-science

    Prepares setup instructions and a named activation file for a user-managed software environment on an Open-Science SSH or Slurm compute host.

    5.5k GitHub stars~2.6k tokensUpdated today
    Research & ScienceAuto-check passed

  • Figure Style

    aipoch/open-science

    Publication-grade correctness and legibility rules for final-deliverable scientific figures, not exploratory plots.

    5.5k GitHub stars~5.1k tokensUpdated today
    Research & ScienceAuto-check passed

  • Add Bactopia Tool

    bactopia/bactopia

    Scaffold a complete Bactopia Tool across all three tiers -- module, subworkflow, and workflow entry point under workflows/bactopia-tools/.

    522 GitHub stars~4.1k tokensUpdated 2 mo ago
    Research & ScienceAuto-check passed

  • Modeling Code and Result Contracts

    yushui2022/MathModel-Skill

    Generates result-evidence contracts, tables and runnable q1 to q3 modeling code scaffolds for a math modeling paper from a model route, a data plan and cleaned data.

    454 GitHub stars~1.4k tokensUpdated 3 days ago
    Research & ScienceAuto-check passed

More from brycewang-stanford/Awesome-Journal-Skills

All 2,387 skills in this repo

  • Aaag Data Analysis

    brycewang-stanford/Awesome-Journal-Skills

    A skill your agent uses when running and reporting the analysis for an Annals of the American Association of Geographers manuscript — spatial statistics and modeling, remote-sensing accuracy, or…

    1.2k GitHub stars~1.3k tokensUpdated 13 days ago
    Auto-check passed

  • Aaag Literature Positioning

    brycewang-stanford/Awesome-Journal-Skills

    A skill your agent uses when positioning an Annals of the American Association of Geographers manuscript in the literature — engaging geographic scholarship across the relevant area and the…

    1.2k GitHub stars~1.3k tokensUpdated 13 days ago
    Auto-check passed

  • Aaag Rebuttal

    brycewang-stanford/Awesome-Journal-Skills

    A skill your agent uses when responding to an Annals of the American Association of Geographers decision letter (major/minor revision) — building a point-by-point response to the subject editor and…

    1.2k GitHub stars~1.4k tokensUpdated 13 days ago
    Auto-check passed

  • Aaag Research Design

    brycewang-stanford/Awesome-Journal-Skills

    A skill your agent uses when defending the research design of an Annals of the American Association of Geographers manuscript — spatial/quantitative analysis and GIScience, remote-sensing and…

    1.2k GitHub stars~1.4k tokensUpdated 13 days ago
    Auto-check passed

  • Aaag Review Process

    brycewang-stanford/Awesome-Journal-Skills

    A skill your agent uses when you need to understand how the Annals of the American Association of Geographers evaluates a manuscript — double-anonymous review routed through a subject editor by…

    1.2k GitHub stars~1.3k tokensUpdated 13 days ago
    Auto-check passed

  • Aaag Submission

    brycewang-stanford/Awesome-Journal-Skills

    A skill your agent uses when running the final pre-submission preflight for the Annals of the American Association of Geographers via ScholarOne Manuscripts — area/article-type selection…

    1.2k GitHub stars~1.6k tokensUpdated 13 days ago
    Auto-check passed

Categories

Questions about Kdd Reproducibility

What does Kdd Reproducibility do?

A skill your agent uses when hardening the reproducibility story of a KDD paper, where reproducibility is an explicit decision factor for area chairs. Kdd Reproducibility is an agent skill from brycewang-stanford/Awesome-Journal-Skills. Use when hardening the reproducibility story of a KDD paper, where reproducibility is an explicit decision factor for area chairs.

When should I use Kdd Reproducibility?

Kdd Reproducibility fits situations like: hardening the reproducibility story of a KDD paper; where reproducibility is an explicit decision factor for area chairs.

How do I install Kdd Reproducibility in Claude Code?

Run `npx skills add brycewang-stanford/Awesome-Journal-Skills --skill kdd-reproducibility -a claude-code`. Or copy the skill folder (KDD-Skills/skills/kdd-reproducibility in brycewang-stanford/Awesome-Journal-Skills) into .claude/skills/kdd-reproducibility in your project. Claude Code loads it when a task matches its description.

How do I install Kdd Reproducibility in Codex?

Run `npx skills add brycewang-stanford/Awesome-Journal-Skills --skill kdd-reproducibility -a codex`. Or copy the skill folder (KDD-Skills/skills/kdd-reproducibility in brycewang-stanford/Awesome-Journal-Skills) into .agents/skills/kdd-reproducibility in your project. Codex loads it when a task matches its description.

Can I use Kdd Reproducibility in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add brycewang-stanford/Awesome-Journal-Skills --skill kdd-reproducibility -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/kdd-reproducibility, .gemini/skills/kdd-reproducibility, .github/skills/kdd-reproducibility and .opencode/skills/kdd-reproducibility in your project.

What does Kdd Reproducibility need to run?

SKILL.md names no scripts, command-line tools or credentials: Kdd Reproducibility is instructions for the agent only. Our summary lists: Python 3.

Does Kdd Reproducibility access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Kdd Reproducibility safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Kdd Reproducibility use?

Kdd Reproducibility is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Kdd Reproducibility use?

About 1.8k tokens (SKILL.md is roughly 7.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Kdd Reproducibility?

Skills that share tags, products or a category with Kdd Reproducibility: Peer Review (K-Dense-AI/claude-scientific-writer, 2.4k stars), CHARLS Paper Reproduction Guide (xjtulyc/MedgeClaw, 617 stars), Compute Environment Setup (aipoch/open-science, 5.5k stars) and Figure Style (aipoch/open-science, 5.5k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Kdd Reproducibility?

brycewang-stanford (a GitHub user) maintains it in brycewang-stanford/Awesome-Journal-Skills, which has 1,231 GitHub stars. The repository holds 2,387 skills in this directory. The repository was last updated on September 27, 2026.

Source: brycewang-stanford/Awesome-Journal-Skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.