Explorar skills

要件やスペックから「あるべきソフトウェア設計」を言語化する。新機能の設計着手時、既存機能の再設計時、仕様の解像度を上げたいときに使う。画面デザイン（UI/UX）は対象外。

Implement AIUC-1 controls in AI systems. Provides concrete technical guidance, implementation patterns, policy templates, and evidence artifact generation for each requirement.

Use when the user explicitly asks for an "AB", adversarial, two-sided, or "red-team" code review of code changes they have made — e.g. "AB review my diff", "do an AB review of my changes", "red-team this code change", "have two reviewers argue about my code". Dispatches two opposing sub-agents — one building the evidence-based case that the change is mergeable, one building the case that it must n

Use when the user explicitly asks to reflect on, review, or learn from the current Claude Code session — e.g. "review this session", "reflect on this session", "why did that take so many tries", "what went wrong this session", or asks to turn this session's mistakes into project rules. Reads the session transcript, finds where Claude's output was rejected or required correction, distills the root

DCB0129 compliance walkthrough for Digital Clinical Safety Officers and health IT manufacturers. Guides you through NHS England's mandatory Clinical Risk Management Standard for the manufacture of health IT systems — hazard identification, risk assessment, safety case, and DCSO responsibilities.

DCB0160 compliance walkthrough for Digital Clinical Safety Officers and NHS/health organisations deploying health IT systems. Guides through NHS England's mandatory Clinical Risk Management Standard for deployment — deployment safety case, integration hazards, local configuration risks, and staff training requirements.

ML/AI health IT safety assessment for Digital Clinical Safety Officers. Applies NHS Digital's supplementary guidance for Assurance of Machine Learning in Autonomous Systems (AMLAS healthcare) alongside DCB0129/DCB0160. Covers ML component identification, training data bias, model performance, explainability, failure modes, and MHRA regulatory considerations. Use when a health IT system includes AI

HTML, HTMX, and CSS best practices following CUBE CSS methodology and Every Layout primitives. Use when working with HTML templates, HTMX interactions, CSS styling, or building server-rendered UI components.

Clinical Safety Case Report (CSCR) writer for health IT systems. Produces a structured safety case argument aligned with NHS England DCB0129/DCB0160 requirements. Covers system description, intended purpose, hazard summary, risk control evidence, the safety argument, and DCSO sign-off sections. Use when preparing for Clinical Authority to Release (CATR) or demonstrating compliance.

Ship a CQAI tool to GitHub and Hugging Face Spaces in a single workflow. Runs pre-flight checks, commits and pushes to GitHub, pushes to HF Spaces, waits for build, and verifies the app is live. Use after /qa-streamlit passes.

Build a complete product experiment plan from a validated hypothesis, metrics, and testing assumptions. Use this skill when the user is ready to prepare an A/B test or structured product experiment for execution.

Design workflows that appropriately balance AI automation with human oversight and intervention. Use this skill when a team is building an AI feature and needs to decide when and how humans stay in the loop.