Skill UI

A comprehensive tool designed to audit the quality, relevance, and redundancy of all Claude skills and commands. It supports two modes: a rapid incremental scan for recently changed skills, and a deep full stocktake using AI judgment and expert checklists to assess viability (Keep, Improve, Update, Retire, Merge).

A structured framework for performing initial security incident triage. It guides analysts through classifying incident types using established standards like NIST SP 800-61r3, assigning business impact severity (P1-P4), enriching alerts with threat intelligence, and determining containment and escalation paths. Use when SIEM/EDR alerts require human judgment or when coordinating multiple concurrent security events.

This skill acts as a critical gatekeeper after a set of experiments are completed. It systematically evaluates raw experimental metrics and outcomes against predefined research claims. The goal is to determine whether the results adequately support a claim, contradict it, or if critical evidence is missing. It guides the research team towards the next decisive action: pivoting the hypothesis, supplementing the data, or confirming the findings. Use this before writing final reports.

This advanced engine creates deeply personalized, relationship-aware daily gifts using a sophisticated five-stage creative pipeline. It first performs an 'Editorial Judgment' to determine if a gift is warranted, and then generates the content concept. Output can be rendered as interactive H5 web pages, AI-generated images, or videos, ensuring that every artifact is highly unique, contextually relevant, and avoids repetition. Ideal for automated journaling, marking milestones, or maintaining thoughtful daily engagement.

StyleSeed is a comprehensive design judgment engine designed to elevate the aesthetic quality of AI-generated user interfaces. It codifies 69 visual rules covering color discipline, spatial rhythm, information hierarchy, and motion feedback. By integrating these principles, developers can ensure that AI output looks professionally designed and polished, rather than simply generated. Includes React components, Tailwind CSS implementation, and actionable design guidelines.

This skill allows users to mine their extensive coding history across various AI agents (Claude, Codex, Gemini, etc.) into a persistent, structured corpus. It surfaces reproducible workflows as evidence and interactively helps the user forge these past practices into reusable, installed skills across their agent ecosystem. It is designed to synthesize judgment from raw transcripts into actionable, portable skills.

This技能 is used when analyzing complex propositions, debating, or reviewing arguments within the CrossFrame Suite. It helps users deconstruct claims into testable arguments, identifying hidden premises, strongest counterarguments, evidence requirements, and conditions for withdrawal, ensuring conclusions are robust and non-judgmental.

This skill is designed for generating structured, insightful, and boundary-aware short answers, such as reader replies, editor responses, or consultation advice. It structures analysis by defining fact boundaries, providing structural judgment, offering necessary critique, and suggesting safe, low-risk next steps. It is primarily activated by the CrossFrame Suite, ensuring the output maintains a canonical Chinese semantic layer while providing professional, actionable advice.

This skill provides a rigorous, multi-layered audit for User Interfaces, proving layout balance and usability beyond subjective judgment. It combines a deterministic spatial audit (calculating centroid, optical-center, and pixel-oracle balance via explicit math) with a professional usability review based on Nielsen's 10 heuristics, conducted by a separate 'fresh-eyes' judge. Ideal for catching misaligned or poorly structured AI-generated designs.

A sophisticated automated loop that uses vision or video models to evaluate subjective aesthetic quality (e.g., timing, color grade, layout). Instead of relying on numeric metrics, the system renders multiple parameterized variants into one artifact, asks the AI to rate them, and suggests optimal values, repeating the process until convergence. Ideal for design iteration where human judgment is the primary success criteria.

This skill generates structured, professional summaries for clinical case presentations, handovers, and educational documentation. It supports established medical formats like SBAR and SOAP notes, ensuring all necessary components—from subjective complaints to recommendations—are systematically included. Designed for healthcare professionals to improve documentation clarity and facilitate seamless patient transitions. Note: This output is strictly for educational and documentation purposes and must not replace professional clinical judgment.

A comprehensive tool designed to audit the quality, relevance, and redundancy of all Claude skills and commands. It supports two modes: a rapid incremental scan for recently changed skills, and a deep full stocktake using AI judgment and expert checklists to assess viability (Keep, Improve, Update, Retire, Merge).