What languages does the delivery-validation skill support?

It automatically detects and supports Node.js (npm), Python (pytest), Rust (cargo), and Go (go test) environments.

What happens if a validation check fails?

The skill returns a structured JSON object including specific fix instructions and the exact commands needed to resolve the failing tests or build errors.

How does it verify that requirements are met?

The skill parses task descriptions for bulleted or numbered lists, extracts them as requirements, and verifies their implementation against the changed files.

Does it prevent regressions?

Yes, it stashes changes to run tests on the previous state and compares the count of passing tests to ensure the new changes haven't broken existing functionality.

Delivery Validation

Name: Delivery Validation
Author: avifenesh

byavifenesh

•

311

セキュリティとテスト

Automates the final verification of code changes by running tests, builds, and requirement checks before shipping.

概要

Delivery Validation is a specialized Claude Code skill designed to ensure production readiness through a rigorous, multi-stage automated checklist. It autonomously detects project environments for JavaScript, Python, Rust, and Go to run relevant test suites and build processes, while also performing intelligent requirement extraction to verify that all task objectives have been met. By comparing test counts to prevent regressions and providing actionable fix instructions upon failure, it streamlines the pre-deployment workflow and maintains high code quality without manual intervention.

主な機能

Automated build verification for production-ready artifacts
Actionable fix instruction generation for failed validation steps
Intelligent requirement extraction and verification from task descriptions
Regression detection via pre-and-post change test count comparison
311 GitHub stars
Multi-language test runner detection (Node.js, Python, Rust, Go)

ユースケース

Pre-deployment health checks to ensure build and test stability
Automating the 'Definition of Done' for AI-generated code tasks
Verifying that all bulleted requirements in a task description are implemented

概要

主な機能

Automated build verification for production-ready artifacts
Actionable fix instruction generation for failed validation steps
Intelligent requirement extraction and verification from task descriptions
Regression detection via pre-and-post change test count comparison
311 GitHub stars
Multi-language test runner detection (Node.js, Python, Rust, Go)

ユースケース

Pre-deployment health checks to ensure build and test stability
Automating the 'Definition of Done' for AI-generated code tasks
Verifying that all bulleted requirements in a task description are implemented