Skill UI

Compares two recorded browser sessions step-by-step by matching steps via shared step IDs. It performs both pixel-level diffing on screenshots (for visual regression) and node-by-node diffing on accessible DOM snapshots (for structural changes). Ideal for visual regression testing, comparing A/B flow variants, and verifying replay consistency in UI development.

Automates the scaffolding of a complete Aggregate Root structure within a specified Bounded Context, adhering strictly to Domain-Driven Design (DDD) principles. It generates necessary files including the entity, supporting value objects, repository interface, domain event definitions, and comprehensive unit test stubs, significantly accelerating the modeling of new business concepts and enforcing domain invariants.

This skill runs an automated evolution process designed to systematically optimize the seven core policy surfaces of an AI agent harness (e.g., planner, contextBuilder, reviewer). By mutating these policies and scoring the variants in a sandbox, it identifies crucial performance improvements without requiring expensive foundation model retraining. It is ideal for empirical configuration tuning and establishing optimized baselines.

This tool provides a comprehensive framework for designing rigorous A/B and multivariate experiments. It guides users through defining hypotheses, calculating statistically necessary sample sizes (handling absolute vs. relative MDE), designing variants, setting metrics, and creating detailed monitoring and stopping rules for CRO experimentation.

Use this tool to pre-screen ad creatives, emails, and landing page copy against synthetic audience panels. Predict response rates, sentiment, and potential objections before launching live campaigns. By scoring variants on criteria like resonance and credibility, it helps marketing teams identify top-performing messages and avoid wasting ad spend on underperforming content.

A structured tool for running A/B tests on marketing content variations. It allows users to compare different approaches—such as headline styles, CTA phrasing, or overall content strategies—by logging variants and evaluating them against technical quality scores (e.g., hallucination, readability). The system provides a comprehensive analysis, identifying a statistically grounded winner and offering actionable recommendations for content improvement.

Generates comprehensive, production-ready video marketing scripts tailored for various platforms (YouTube, TikTok, Reels, etc.). It strategically integrates multiple hook variants, detailed timestamps, visual direction notes, optimized CTA placements, and A/B test concepts for thumbnails, ensuring the final content is highly engaging and maximizes conversion rates.

A comprehensive guide for authors assessing whether their genomics or genetics research meets the rigorous standards of Nature Genetics. It details the journal's scope (GWAS, functional genomics, rare variants), the required methodology and evidence bar, and essential house style guidelines for submission.

This skill transforms raw genomic identifiers (VCF, HGVS, rsID) into highly annotated, consequence-predicted records. It utilizes powerful, open-source tools like Ensembl VEP and gnomAD to predict variant consequences (e.g., missense, nonsense), map coordinates, and attach population allele frequencies. Furthermore, it links these molecular findings to clinical and phenotype context extracted by OpenMed, forming a comprehensive research record.