Skill UI

A comprehensive security assessment skill for identifying vulnerabilities in federated identity systems, including SAML, OIDC, and OAuth 2.0 flows. It covers common misconfigurations, such as signature tampering, PKCE failures, redirect URI issues, and token confusion, ensuring robust SSO implementation.

This module converts diverse clinical inputs—including scanned images, faxes, CSV/TSV exports, and C-CDA XML—into a normalized, clean text format (`ExtractedDocument`). It supports on-device OCR (Tesseract/PaddleOCR) and structured redaction for tables. It is the crucial first stage for feeding de-identification and Named Entity Recognition (NER) pipelines.

A comprehensive, multi-phase methodology for debugging complex software issues. This guide mandates systematic root cause investigation, starting from symptom collection, code path tracing, hypothesis formulation, and deterministic reproduction. It teaches developers to fix the underlying cause, not just the visible symptoms, ensuring highly robust and minimal changes.

A comprehensive, systematic workflow for reverse engineering client-side JavaScript. It guides users through five stages: Observation, Capture, Rebuilding, Patching, and Deep Dive. Applicable for locating API signatures, analyzing encryption parameters, tracking runtime function calls, and reconstructing complex web environments locally for evidence collection and reproduction. Utilizes advanced tools like jshookmcp and various js-reverse_* utilities.

This extension provides comprehensive management for external access using the standard Kubernetes Gateway API and Traefik as the underlying proxy. It supports multi-tenant scoping across cluster, workspace, and namespace levels, enabling robust, modern API gateway configuration within the KubeSphere ecosystem.

This skill generates complete, deploy-ready static landing pages by analyzing a source code repository. It reads key files (README.md, CHANGELOG.md, etc.), extracts structured content (features, installation guides, FAQ), and compiles it into a professional, responsive HTML/CSS/JS site. Ideal for standardizing open-source project showcases, converting technical documentation into marketing pages, or quickly building a project homepage.

A comprehensive tool designed to help developers compare multiple technical approaches for solving a single problem. It systematically identifies and analyzes the trade-offs, complexity, performance, and learning curves associated with different design patterns (e.g., state management, data fetching, API design). Ideal for architectural reviews and making informed, robust technology stack decisions.

This tool tests your deep understanding of code you have recently written or reviewed. It generates structured, multi-level questions (Recall, Understanding, Application, Analysis, Synthesis) based on the provided code. Users can focus the quiz on specific areas like error handling, performance, or security, making it an excellent mechanism for continuous learning and solidifying technical knowledge.

This tool provides a structured, multi-stage method for learning any technical concept, from basic definitions to advanced implementation details and common pitfalls. It progressively deepens the explanation, making complex topics easy to grasp.

This skill resolves entities extracted by OpenMed (e.g., Disease, Drug) to standardized Concept Unique Identifiers (CUIs) within the UMLS Metathesaurus. It is essential for cross-vocabulary normalization, allowing users to unify concepts (like SNOMED CT, ICD-10, RxNorm) to a single canonical CUI. Crucially, the process relies entirely on the user's own UTS API key, ensuring that the Metathesaurus content is never bundled or cached.

A comprehensive framework for conducting authorized security assessments on Large Language Models (LLMs) and AI agents. It covers critical vulnerabilities including advanced prompt injection (OWASP Top 10), tool abuse, RAG data poisoning, memory manipulation, and system prompt extraction. Ideal for red teaming, security auditing, and ensuring the robustness of AI systems in production environments.

This skill manages the loading and efficient caching of OpenMed clinical/biomedical Named Entity Recognition (NER) models. Models are downloaded once from the Hugging Face Hub, enabling full offline operation afterward. It supports model reuse across multiple calls, configuring cache directories, specifying devices (CPU/GPU), and handling various model identifiers (registry keys, full HF IDs, or local paths).