Does it check if IDs match across different files?

Yes, it includes a specialized cross-reference function that ensures, for example, that a Flow ID references a valid PRD feature.

How does it handle validation errors?

It provides structured JSON output detailing the specific field that failed, the reason for the error (like pattern mismatch), and the count of total errors found.

Can I use this for early-stage drafting?

Absolutely. It supports a 'Draft Mode' which provides minimal validation for work-in-progress files, alongside 'Strict' and 'Lenient' modes.

Which artifact types does this skill support?

This skill validates a wide range of artifacts including PRDs, Flows, ERDs, Journeys, Tasks, ADRs, and Scaffolds.

Pipeline Data Validator

Name: Pipeline Data Validator
Author: KreativReason

byKreativReason

0•

Security & Testing

Ensures data integrity across the development lifecycle by validating JSON artifacts against strict Pydantic schemas.

The Pipeline Data Validator skill provides a robust framework for verifying the integrity and consistency of technical artifacts such as PRDs, ERDs, and architectural decision records. By leveraging Pydantic schemas, it automates the validation of complex JSON structures, enforces standardized ID patterns (e.g., FR-XXX for features), and performs cross-reference checks to ensure that flows and tasks correctly map to their parent requirements. This skill is essential for maintaining a high-quality, 'single source of truth' throughout an AI-driven development pipeline, offering flexible validation modes ranging from lenient drafts to strict production-ready checks.

Key Features

01Cross-reference checking to ensure consistency between related documents

02Support for multiple validation modes: Strict, Lenient, and Draft

03Automated schema validation for PRD, Flow, ERD, and ADR artifacts

040 GitHub stars

05Strict enforcement of ID naming conventions and unique identifier rules

06Real-time validation feedback with detailed error reporting for developers

Use Cases

01Refining work-in-progress JSON drafts into production-grade technical specifications

02Verifying project documentation consistency before triggering code generation agents

03Automating quality gates in CI/CD pipelines for architectural and design artifacts

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add kreativreason/merged-end-to-end-ai-dpp---e2e-cli validation

For use in Claude.ai and ChatGPT

Download Skill