Can it check GitHub Actions status?

Yes, it includes processes to audit GitHub Actions workflow runs via the GitHub CLI to ensure CI passes before any certification is issued.

Does it replace a human QA engineer?

No, it acts as a specialist auditor that verifies the work of both developers and QA agents against the source of truth, specifically challenging inflated quality ratings.

Which tech stacks does this skill support?

While it can be adapted, it defaults to Corey's stack: Expo (React Native), Supabase (Edge Functions/DB), and Vercel (web previews), utilizing tools like Detox and Playwright.

Why does it default to 'NEEDS WORK'?

By defaulting to a failing state, the skill ensures a high quality bar. It places the burden of proof on the code and automated tests rather than assuming success.

What is a 'fantasy approval'?

A fantasy approval occurs when a feature or build is marked as 'complete' or 'A+' without sufficient proof that it actually works across all target environments. This skill prevents them by requiring logs and test output.

Production Reality Checker

Name: Production Reality Checker
Author: coreymaypray

bycoreymaypray

0•

Security & Testing

Enforces evidence-based production certification by auditing system health, cross-validating QA findings, and preventing premature release approvals.

Reality Checker serves as the final line of defense against unrealistic project assessments and 'fantasy approvals.' It automatically defaults to a 'NEEDS WORK' status, requiring overwhelming evidence—such as logs, test results, and visual captures—before certifying a build for production. Designed for rigorous environments using Expo, Supabase, and Vercel, it cross-references documentation against actual source code and CI results to ensure that features aren't just claimed, but are fully functional and ready for real-world deployment.

Key Features

01Evidence-based deployment readiness assessments

02Automatic fail triggers for broken CI, missing source, or failing E2E flows

030 GitHub stars

04Strict quality rating system (C+ to A-) to prevent inflated assessments

05Automated cross-validation of QA findings against implementation

06End-to-end integration auditing for mobile and web journeys

Use Cases

01Auditing developer contributions for technical debt and spec compliance

02Pre-release production certification for mobile and web applications

03Validating Supabase Edge Functions and backend logic against logs

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add coreymaypray/sloth-skill-tree reality-checker

For use in Claude.ai and ChatGPT

Key Features

01Evidence-based deployment readiness assessments

02Automatic fail triggers for broken CI, missing source, or failing E2E flows

030 GitHub stars

04Strict quality rating system (C+ to A-) to prevent inflated assessments

05Automated cross-validation of QA findings against implementation

06End-to-end integration auditing for mobile and web journeys

Use Cases

01Auditing developer contributions for technical debt and spec compliance

02Pre-release production certification for mobile and web applications

03Validating Supabase Edge Functions and backend logic against logs

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add coreymaypray/sloth-skill-tree reality-checker

For use in Claude.ai and ChatGPT