skills.rest
JunMA98
JunMA98COMMUNITY
Explore all skills by this author →
GitHub

code-reproduction

Reproduce reported results with a structured plan and execution trace.

3|Updated Mar 11, 2026
Search Tags:#research#reproducibility#reproduction#experiments#environment-setup#discrepancy
One-click install
npx skills add https://github.com/JunMA98/Computer-science-claude-skills --skill code-reproduction
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: code-reproduction
Source: https://github.com/JunMA98/Computer-science-claude-skills/tree/main/skills/code-reproduction
Command: npx skills add https://github.com/JunMA98/Computer-science-claude-skills --skill code-reproduction

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Reproduce reported results with a clear, auditable process that avoids overstating outcomes and documents every discrepancy.

Core Features & Use Cases

  • Reproduce stated results from papers or repos with a structured plan and execution trace.
  • Generate environment reconstructions, minimum-viable-path execution, and discrepancy reports to guide verification.
  • Track reproduction status (exact, partial, failed, or blocked) and surface undocumented assumptions for review.

Quick Start

Create a reproducibility plan for the target paper or repository and begin the minimum viable path to reproduce the results.

Frequently Asked Questions about code-reproduction

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I reproduce experiments from an academic paper and verify the reported results?▼

Reproduce experiments by creating a structured plan that guides environment reconstruction, minimum-viable-path execution, and discrepancy logging to verify reported results with an auditable execution trace.

What is the best way to track reproduction status and log discrepancies during environment setup?▼

Track reproduction status by logging discrepancies during environment reconstruction and reporting outcomes as exact, partial, failed, or blocked while surfacing undocumented assumptions for review.

Can I use this workflow to reproduce results from public benchmark repositories without overstating outcomes?▼

Yes, you can reproduce results from public benchmark repositories using an auditable process that documents every discrepancy and reports explicit reproduction status to avoid overstating outcomes.

Why does my experiment reproduction fail due to undocumented assumptions in the original repository?▼

Experiment reproduction fails when undocumented assumptions exist, which this process surfaces for review by enforcing environment reconstruction and discrepancy logging to identify missing repository requirements.

Do I need a structured reproduction plan to verify results from repository-based experiments?▼

Yes, a structured reproduction plan is needed to guide environment reconstruction, extract claims, and execute minimum-viable-path verification workflows for repository-based experiments.

What does a minimum-viable-path execution mean for reproducing reported results?▼

Minimum-viable-path execution means following the most direct structured workflow to reproduce reported results while enforcing environment reconstruction and discrepancy logging throughout the verification process.

Related Skills

Explore More Education & Research Skills→
Lhhiep-maxcodeLhhiep-maxcode

nemotron-nano3

Answers factual questions about Nemotron 3 Nano architecture, training, evaluation, and deployment from authoritative sources.

Community
Advanced
BoomberAspBoomberAsp
1

baseline

Attach, import, reproduce, repair, verify, compare, or publish baselines and metrics.

Community
Advanced
giosuetedeschi-specgiosuetedeschi-spec

langfuse

Track LLM traces, spans, and metrics for production AI applications.

Community
Advanced
tizeetizee
2

uv-script

Create and run standalone Python scripts with inline dependency metadata using uv.

Community
Advanced