Skill UI

Security hardening guidance for AI agent frameworks processing untrusted content, invoking tools, writing files, managing identifiers, or handling credentials. Use when building or reviewing agent runtimes, autonomous workers, tool gateways, memory services, or multi-tenant deployments.

End-to-end organization of Power BI or Analysis Services semantic models to be report-ready: table groups, display folders, Title Case naming conventions, hierarchies, sort-by columns, hiding, descriptions, format strings, perspectives, translations and security roles.

This skill activates a cheeky Cockney persona for coding tasks. It encourages terse replies, minimal diffs, and efficient code changes. Use it when you want to avoid bloat or over-engineering. The AI will speak with a London accent slang (lite, full, guvnor modes) while maintaining security standards. Turn off with specific commands.

This skill covers creating a new Quarkus project using the Quarkus CLI, Maven plugin, or Gradle. It includes selecting appropriate extensions for REST APIs, databases, and security. Users learn to scaffold applications, configure build tools, and run in dev mode for live reload. Suitable for Java backend development and microservices initialization.

Write Quarkus application code including REST APIs, validation, OpenAPI, CDI, Hibernate ORM, and reactive messaging. Configure extensions, manage security with OIDC, and implement observability features like health checks and metrics. Use this skill for implementing or modifying features in existing Quarkus projects, ensuring proper build-time resolution and virtual thread usage.

This skill enables users to generate AI audiobooks from manuscripts or course text using the Beatra service. It supports voice cloning and text-to-speech narration. The package is pinned and verified for security. Users must authorize access before use. It works within Claude Code environments.

This skill statically inspects local JavaScript, Electron, native binary, .NET, and CUDA applications to explain feature functionality with file and line evidence. It supports analyzing packaged .asar files, ELF/PE binaries, EVM bytecode, and CUDA programs without execution. Users must be authorized to study the target. It helps understand or reimplement features safely by parsing code and metadata, refusing requests to bypass security mechanisms or DRM.

Commit Craft is a Git workflow tool that organizes mixed changes into logical, reviewable commits before pushing. It blocks secrets like .env files, lints commit messages against standards, and ensures each commit builds independently. Using Python scripts, it plans series, gates secrets, and validates messages without modifying the repository directly, enhancing code quality and security.