Verified:GitHub

validation-pipeline

Build multi-stage validation pipelines with hallucination detection and citation verification.

Updated Aug 23, 2026

One-click install

npx skills add https://github.com/jennifer-mckinney/my-skills --skill validation-pipeline

Or copy as Structured Prompt for Agent▼

Please help me install this Agent Skill.
Skill: validation-pipeline
Source: https://github.com/jennifer-mckinney/my-skills/tree/main/validation-pipeline
Command: npx skills add https://github.com/jennifer-mckinney/my-skills --skill validation-pipeline

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires transformers, torch, sentence-transformers, scipy, scikit-learn, matplotlib, dateparser, cryptography, PyYAML, spacy, pytest, and includes scripts (resource) and references (resource) and templates (resource) components.

What problem does it solve?

Building AI systems that meet stringent accuracy, reliability, and regulatory compliance (like FDA/EMA) is incredibly complex and error-prone. This Skill provides a comprehensive framework to construct multi-stage validation pipelines, ensuring zero-hallucination, verifiable citations, and high confidence in AI outputs.

Core Features & Use Cases

Multi-Stage Validation: Implement pre-generation data quality checks, real-time confidence monitoring, and post-generation citation, hallucination, and consistency checks.
Hallucination & Citation Detection: Automatically detect fabricated claims and verify that all information is accurately sourced.
Confidence Scoring: Quantify AI output quality with multi-dimensional confidence scores for critical decision-making.
Use Case: A healthcare AI system needs to provide clinical decision support with 99.9%+ accuracy. Use this Skill to build a validation pipeline that checks input data quality, ensures every claim is cited, detects any hallucination, and provides an audit trail for FDA compliance.

Quick Start

Use the validation-pipeline skill to detect hallucinations in 'response.json' using 'sources.json' with a 99.9% confidence threshold.