consistency-auditor

Audit cross-media consistency across tex, code, results, data, and appendix.

452|24|Updated Apr 26, 2026
One-click install
npx skills add https://github.com/zhnnky329/MathModeling-skills --skill consistency-auditor-zhnnky329
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: consistency-auditor
Source: https://github.com/zhnnky329/MathModeling-skills/tree/main/.claude/skills/consistency-auditor
Command: npx skills add https://github.com/zhnnky329/MathModeling-skills --skill consistency-auditor-zhnnky329

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Audit cross-media consistency across tex/code/results/data/appendix to identify divergences between claims and canonical sources. It operates as an independent audit layer that does not trust any single skill's self-declaration of "done".

Core Features & Use Cases

  • Independent cross-media verification across paper drafts, code, and results.
  • Five-dimension audit coverage: numbers, file names, symbols, parameters, and decision provenance.
  • Traces every judgment to a DECIDED record, surfacing divergences that block final assembly.

Quick Start

Run the auditor after drafting sections to identify divergences between claims and canonical sources.

Frequently Asked Questions about consistency-auditor

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

FAQPage Schema
How do I audit cross-media consistency between my paper draft and source code?

Cross-media consistency auditing checks tex, code, results, data, and appendix files to identify divergences between claims and canonical sources across five dimensions: numbers, file names, symbols, parameters, and decision provenance.

What is cross-media inconsistency in research artifacts?

Cross-media inconsistency occurs when numerical claims, figure references, symbols, or parameters in a paper draft diverge from canonical sources in the code, data, or appendix. An independent audit layer traces these mismatches to flag blocking divergences for repair.

How do I trace numbers and parameters across subquestions and repository artifacts?

Tracing numbers and parameters applies an audit across all subquestions and repository artifacts to enforce traceability from every judgment to DECIDED records, surfacing divergences where claims do not match canonical sources for repair.

When do I need to run a consistency audit on my research repository?

Run a consistency audit after drafting sections to identify divergences between claims and canonical sources. It operates as an independent verification layer that does not trust any single skill's self-declaration of done, ensuring final assembly is not blocked.

Does the cross-media audit work without relying on self-declared completion statuses?

The cross-media audit operates as an independent layer that does not trust any single skill's self-declaration of done. It independently verifies numbers, symbols, parameters, and decision provenance across tex, code, and data files to enforce traceability.

What are the limitations of cross-media consistency auditing for research papers?

Cross-media consistency auditing limits its scope to five dimensions—numbers, file names, symbols, parameters, and decision provenance—across tex, code, results, data, and appendix. It flags blocking divergences for repair but does not automatically resolve the mismatches.