cross-test

Adversarially validate retikz implementation behavior against existing contracts and tests.

12|2|Updated Nov 27, 2024
One-click install
npx skills add https://github.com/Pionpill/retikz --skill cross-test
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cross-test
Source: https://github.com/Pionpill/retikz/tree/main/.agents/skills/cross-test
Command: npx skills add https://github.com/Pionpill/retikz --skill cross-test

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you discover real defects by attacking the current implementation with edge cases derived from existing code, public contracts, schemas, and tests—without inventing requirements.

Core Features & Use Cases

  • Edge-case discovery from reality: Builds an attack map from packages/core and packages/react implementation, IR/schema, and existing tests.
  • User-impact classification: Separates findings into BLOCKING/WARNING/INFO with evidence, user diagnostics, and follow-up actions.
  • Regression hardening: Turns confirmed bugs into new automated tests that are expected to fail on current code, then become durable regression coverage.

Quick Start

Run cross-test on packages/core first, targeting an IR schema/parser or compile path, then add a minimal failing edge-case test and finish with a categorized report.

Frequently Asked Questions about cross-test

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

FAQPage Schema
How do I find edge-case bugs in my React rendering and IR schema parsing pathways?

Adversarial cross testing uncovers defects by attacking your current implementation with edge cases derived from existing code, public contracts, schemas, and tests. It strictly derives test cases from reality without inventing requirements.

What is the best way to generate regression tests for discovered bugs?

Regression hardening turns confirmed bugs into new automated tests that are expected to fail on current code, then become durable regression coverage. You run minimal relevant vitest commands to validate these edge cases.

How does user-impact classification work when auditing core compile pathways?

User-impact classification separates findings into BLOCKING, WARNING, and INFO categories with concrete evidence, user diagnostics, and follow-up actions. This structured report helps prioritize bug fixes based on severity.

Can I use adversarial testing to validate a specific TODO or bug hypothesis?

Yes, cross testing applies to auditing a single package or module, or targeting a specific TODO, ADR, or bug hypothesis across core IR parsing and compile pathways. It builds an attack map from implementation details.

Do I need existing tests to run adversarial cross testing on my packages?

Cross testing builds an attack map from your implementation, IR schema, and existing tests. While it uses current tests as a baseline for deriving edge cases, it strictly validates behavior against public API and schema contracts.