Skill UI

Handles encryption and decryption tasks via the encryption-tool plugin to protect sensitive information, letting Claude choose AES, RSA, or other algorithms, prompt for keys when needed, and optionally save encrypted or decrypted files.

This guide provides comprehensive best practices for securing Evernote API integrations. It covers essential topics like secure credential management using environment variables or vaults, implementing CSRF-protected OAuth flows, encrypting access tokens at rest using AES-256-GCM, rigorous input validation, and ensuring secure logging practices by redacting sensitive data. Use this when hardening your application against data breaches.

This skill provides AI-powered capabilities for advanced image manipulation. It specializes in style transfer, allowing users to apply the aesthetic characteristics of one image to another, and object removal, enabling seamless retouching by eliminating unwanted elements. It is essential for designers, photographers, and artists requiring high-quality, creative photo editing and visual enhancement.

Provides structured guidance for distinctive, intentional visual design of frontend interfaces. Helps break generic AI-style patterns by grounding choices in subject matter, typography, and visual structure. Useful for designers and developers building new UI or reshaping existing interfaces with a unique aesthetic direction.

Frontend Slides Designer generates zero-dependency, animation-rich HTML presentations from scratch or by converting PowerPoint decks, ensuring viewport-fitted slides with distinctive aesthetics for pitch decks, tutorials, or showcases.

A comprehensive reference architecture for deploying meeting intelligence platforms (like Granola) across large organizations. This blueprint details the core data capture pipeline, integration topology using middleware (Zapier), multi-workspace structure (Sales, HR, Engineering), and robust security layers (SSO, AES-256) to ensure automated workflows from meeting to action item across all departments.

This comprehensive guide outlines best practices for securing Granola meeting data. It covers critical compliance standards, including SOC 2 and GDPR, and details technical controls such as AES-256 encryption, TLS 1.3 transmission, and robust configuration of SSO and data retention policies for enterprise use.

This skill provides comprehensive guidelines for handling sensitive people datasets from Juicebox AI. It covers critical data classification (PII, search results), required technical controls (AES-256 encryption, field-level masking), and adherence to international privacy regulations like GDPR and CCPA. It ensures users implement lawful data processing, proper consent tracking, and secure data export mechanisms.

This guide details how to apply sophisticated artistic styles and visual effects to Kling AI video generation. Users can achieve diverse aesthetics—such as Cinematic, Anime, Watercolor, or Cyberpunk—either through detailed prompt engineering or by utilizing the dedicated Effects API (e.g., hug, kiss). It also provides methods for pre-processing images using Kolors for highly consistent and professional video outputs.

This guide provides a comprehensive setup for Replit applications. It covers configuring essential files like .replit and replit.nix for dependency management, securely managing environment secrets (AES-256), and integrating robust user authentication via Replit Auth (supporting Google, GitHub, etc.). Ideal for starting a new, production-ready Replit project.

A comprehensive guide detailing security best practices for building and deploying applications on the Replit platform. Learn how to properly manage sensitive data using Replit Secrets (AES-256 encryption), implement secure service-to-service authentication via REPL_IDENTITY tokens, validate custom headers, and prevent common vulnerabilities like hardcoded API keys and SQL injection. Essential for auditing and securing production environments.

SAELens provides a framework for training and analyzing Sparse Autoencoders (SAEs). SAEs decompose the dense, often polysemantic activations of large language models into sparse, monosemantic features. Use this when you need to discover the discrete, interpretable concepts a model has learned, study feature superposition, or analyze specific safety-relevant behaviors (like bias or deception) within deep neural networks.