verification

Run deterministic verification commands before asserting research or writing success.

Updated May 15, 2026
One-click install
npx skills add https://github.com/lth0/codexSkill --skill verification-lth0
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: verification
Source: https://github.com/lth0/codexSkill/tree/main/skills/agents/research-writing/verification
Command: npx skills add https://github.com/lth0/codexSkill --skill verification-lth0

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents misleading “done” or “verified” claims by forcing you to run concrete verification commands and read the full output before you assert success.

Core Features & Use Cases

  • Verification gating for academic/research work: Guards statements like “chapter finished,” “references verified,” or “quality gate passed” until evidence is checked.
  • Evidence-driven validation patterns: Defines repeatable checks for file existence, word counts, citation authenticity (e.g., DOI/CrossRef), argument support via evidence maps, and formal formatting/quality gates.
  • Red-flag stop rules: Instructs you to stop when you see “should/might/look like” phrasing, premature satisfaction, or any claim made without executed validation steps.

Quick Start

Use the verification skill when you are about to say a chapter is complete or that citations are verified, and ensure you run the relevant verification commands and paste the full outputs before making any success claim.

Frequently Asked Questions about verification

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

FAQPage Schema
How do I verify citations and check formatting before claiming a research chapter is complete?

To verify citations and formatting, run deterministic verification steps like checking DOI authenticity and format compliance, then read the complete command output before asserting chapter completion.

What is evidence-based verification in academic research writing?

Evidence-based verification in academic research writing requires executing concrete validation commands and reviewing their full outputs to substantiate success claims, preventing misleading statements about finished chapters or verified references.

How do I validate literature search results to ensure academic integrity?

Validate literature search results by running repeatable checks for citation authenticity through DOI or CrossRef validation and mapping arguments to evidence, ensuring all claims are directly supported by executed validation steps.

Does automated quality gate checking prevent unsubstantiated success claims in research workflows?

Automated quality gate checking prevents unsubstantiated success claims by enforcing red-flag stop rules that halt progress when detecting premature satisfaction or unverified statements like "should" or "might" without executed validation.

When should I stop and run verification commands during research writing?

You should stop and run verification commands immediately before making any success claim, especially when you see "should", "might", or "look like" phrasing, or when attempting to assert a chapter is finished or references are verified.