Verify

Validate code changes against predefined acceptance criteria using test evidence and LLM judgements.

6|3|Updated Apr 22, 2026
One-click install
npx skills add https://github.com/lordmacu/nexo-rs --skill verify-lordmacu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Verify
Source: https://github.com/lordmacu/nexo-rs/tree/main/skills/verify
Command: npx skills add https://github.com/lordmacu/nexo-rs --skill verify-lordmacu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill allows you to validate that changes in your codebase actually meet predefined acceptance criteria, providing evidence-based outcomes to ensure changes do not break existing functionality.

Core Features & Use Cases

  • Bounded Acceptance Verification: Runs checks and an explicit LLM judge over evidence.
  • Automated Testing: Uses a predefined set of checks to ensure criteria are met.
  • Use Case: When a pull request is created and the developer needs to validate that the change adheres to acceptance criteria without introducing regressions.

Quick Start

Run the 'Verify' Skill on your latest change by providing the acceptance criteria and the code modifications.

Frequently Asked Questions about Verify

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

FAQPage Schema
How do I validate code changes against predefined acceptance criteria?

Acceptance testing for pull requests works by executing predefined checks against your codebase modifications and collecting evidence from test results. An LLM judge then evaluates this evidence against your predefined acceptance criteria to provide an evidence-based decision on whether the change adheres to quality standards.

How do I automate change validation for a pull request?

To automate change validation, provide your user-defined acceptance criteria and optional candidate testing commands. The verification process runs these checks, collects the evidence, and uses an LLM judge to determine if the code modifications meet the specified standards, preventing regressions.

What is LLM judgment in software quality assurance?

LLM judgment in software quality assurance is an evaluation mechanism where a language model acts as an explicit judge over collected test evidence. It determines whether code changes satisfy predefined acceptance criteria, providing an evidence-based outcome for change validation.

What do I need to provide for bounded acceptance verification?

For bounded acceptance verification, you need to provide user-defined acceptance criteria and optional candidate commands for testing. The system uses these inputs to execute checks, gather evidence, and apply an LLM judge to validate that your code changes meet the required standards.

Does code verification with an LLM judge require predefined test commands?

No, code verification with an LLM judge does not strictly require predefined test commands; candidate commands for testing are optional. However, you must provide predefined acceptance criteria so the system has a baseline to evaluate the collected evidence against.