
Qing dynasty court roleplay mode. Activates imperial court persona for all agent responses. Use when user wants to enable court roleplay, mentions qing, or wants a fun imperial Chinese tone.
Create OpenAPI specs, SDK docs, and API documentation. Use for API documentation or client library generation.
Design APIs, microservices, and database schemas. Use for server-side development, API design, or system architecture.
Use when auditing a case report or case series manuscript against the CARE checklist. Triggers on case reports, rare disease presentations, unusual clinical findings, or single-patient studies.
Design AWS/Azure/GCP infrastructure, implement IaC, and optimize costs. Use for cloud architecture, cost optimization, or migration.
Use when auditing a randomized controlled trial manuscript against the CONSORT 2025 checklist. Triggers on RCTs, randomized trials, clinical trials, intervention studies with randomization, parallel-group designs, crossover trials, or factorial designs.
Data analysis, SQL queries, BigQuery operations, and data insights. Use for data analysis tasks and queries.
Configure CI/CD, Docker, and cloud deployments. Use for deployment setup, containers, or infrastructure automation.
Use when polishing or reviewing English text to avoid AI-flavored vocabulary. Scans for LLM-preferred words identified by PubMed research (delve, meticulous, underscore, etc.) and suggests natural alternatives.
Write modern JavaScript with ES6+, async patterns, and Node.js. Use for JS development, async debugging, or optimization.
Assemble publication-ready meta-analysis manuscripts with tables, figures, and references. Use when completing systematic review/meta-analysis manuscripts for journal submission.
Play Texas Hold'em poker as an AI agent. Use when joining a poker game, making betting decisions, or analyzing opponents.
Generate Quarto Book project structure with chapters, configuration, and output settings. Use when user wants to create a book, multi-chapter document, technical manual, or asks about Quarto book setup.
Write idiomatic Rust with ownership, lifetimes, and traits. Use for Rust development, memory safety, or systems programming.
Assemble multi-panel scientific figures with panel labels (A, B, C) at publication quality (300 DPI) using R. Use when combining individual plots into journal-ready figures.
Security audits, vulnerability detection, and secure coding. Use for security reviews, auth implementation, or OWASP compliance.
Write SQL queries, optimize execution plans, and design schemas. Use for query optimization, complex joins, or database design.
Use when auditing a prediction model or clinical AI manuscript against the TRIPOD+AI checklist. Triggers on prediction models, prognostic scores, diagnostic models, machine learning clinical tools, risk calculators, AUC/c-statistic reporting, or AI-assisted clinical decision support.
Create new Claude Code skills with proper structure. Use when user wants to add a skill, create a slash command, or automate a workflow. Fetches official docs and guides through skill creation.
Refactor code for quality, reduce technical debt, and improve maintainability. Use for cleanup tasks and code improvements.
Debug errors, test failures, and unexpected behavior with log analysis and correlation. Use when encountering issues, error messages, analyzing logs, or investigating production errors.
Load context for a new agent session by analyzing codebase structure and README. Use when starting a new session or needing to understand an unfamiliar codebase.
Publish a skill folder as a .skill package to GitHub. Use when the user wants to publish, release, or distribute a skill folder — creates a public GitHub repo with description, README with badges, .gitignore, and a GitHub Action that zips the skill folder into a .skill artifact and creates releases tagged by commit SHA. Trigger on: "publish skill", "release skill", "make skill public", "distribute skill", "package skill for GitHub", or when user has a SKILL.md folder ready for distribution.
Update dev docs before session compaction to capture progress. Use before compacting conversation, ending sessions, or switching tasks.
Create test suites with unit, integration, and e2e tests. Use when setting up tests, improving coverage, or when user asks about testing.
Generate semantic codebase symbol map for precise code navigation. Use when starting work on unfamiliar codebase, before refactoring, or when you need to understand where classes/functions/interfaces are defined instead of using rg text-matching.
Build LLM applications, RAG systems, and prompt pipelines. Use for LLM features, chatbots, or AI-powered applications.
List All Tools. Use when user wants to see available tools or needs a tool reference.
Design system architecture, APIs, and component interfaces. Use for architectural decisions and system design.
Write efficient C code with proper memory management and system calls. Use for C optimization, memory issues, or system programming.
Run a comprehensive code review
Write modern C++ with RAII, smart pointers, and STL. Use for C++ development, memory safety, or performance optimization.
Build ETL pipelines, data warehouses, and streaming architectures. Use for data pipeline design or analytics infrastructure.
Database optimization, query tuning, migrations, and administration. Use for database performance issues, schema design, or operational tasks.
Create comprehensive strategic plan and dev docs for large tasks. Invoke when starting large tasks that span multiple sessions.
Scan the current environment for developer experience readiness — tools, runtimes, permissions, network, and container config. Use when entering a container, checking what's installed, auditing dev tools, or diagnosing missing dependencies.
Build React components, responsive layouts, and handle state management. Use for UI development, styling, or frontend performance.
Understand the current state of the git repository. Use when user wants to check git status or before making commits.
Write idiomatic Go with goroutines, channels, and interfaces. Use for Go development, concurrency, or performance.
Handle production incidents with urgency. Use when production issues occur for debugging, fixes, and post-mortems.
Guide medical manuscript writing with argumentation-driven structure, section-by-section best practices, and sentence-level craft. Use when writing, revising, or reviewing original research manuscripts for biomedical journals, or when a draft reads flat or reviewer comments cite structural issues.
Implement ML pipelines, model serving, and feature engineering. Use for ML model integration or production deployment.
Debug network connectivity, configure load balancers, and analyze traffic. Use for connectivity issues, network optimization, or protocol debugging.
Use when auditing a systematic review or meta-analysis manuscript against the PRISMA 2020 checklist. Triggers on systematic reviews, meta-analyses, pooled analyses, forest plots, search strategies, or study selection flow diagrams.
Optimize prompts for LLMs and AI systems. Use when building AI features, improving agent performance, or crafting system prompts.
Write idiomatic Python with advanced features. Use for Python development, refactoring, or optimization.
Generate TypeScript types from JSON files using quicktype CLI. Use when user wants to create types/interfaces from JSON data, API responses, or config files.
Review the current Claude Code session conversation. Extracts only human-readable content (user prompts and agent prose responses, skipping tool calls and tool results), then produces a structured retrospective report covering accomplishments, efficiency improvements, English corrections, learning recommendations, and CLAUDE.md suggestions. Use when the user asks to review, retrospect, or summarize the session — especially at the end of a work session.
Use when auditing an observational study manuscript (cohort, case-control, or cross-sectional) against the STROBE checklist. Triggers on retrospective studies, hospital data, registry/database studies like TCGA, SEER, NHIRD, or any non-randomized clinical research.
Expert web research with advanced search techniques. Use for deep research, information gathering, or fact verification.
Generates new Claude Code skills with proper structure and best practices. Use when user asks to create a new skill, slash command, or wants to automate a workflow. Guides through requirements gathering, structure generation, and validation.
Create and present web-based slides for developers using Markdown, Vue components, code highlighting, animations, and interactive features. Use when building technical presentations, conference talks, or teaching materials.
Runs Vale prose linter on markdown/text files and auto-fixes issues. Use when the user asks to lint, proofread, or improve writing quality of markdown or text files.
Use when reviewing a data visualization or figure for clarity, checking if a graph communicates its message without additional context, or iterating on R/Python plot scripts until a naive reader can fully understand the figure.
Converts narrative medical text into Pocket Medicine bullet-style notes with proper abbreviations, then modularizes sections exceeding 20 lines into linked standalone files.
Improve developer experience, tooling, and workflows. Use when setting up projects or reducing development friction.
Profile applications, optimize bottlenecks, and implement caching. Use for performance issues or optimization tasks.
Manage Tailscale Funnel routes for multi-service coexistence. Use when checking funnel status, adding/removing service routes, diagnosing tunnel connectivity, or when "funnel", "tailscale", "tunnel", "port routing" is mentioned. Also use proactively when deploying new services that need external HTTPS access.
Hook control - enable, disable, and manage Claude Code hooks. Use when you need to manage, debug, or monitor Claude Code hooks for the current session.
Analyze Claude Code context efficiency for current project. Invoke when evaluating session efficiency or optimizing Claude Code usage.
Git checkpoint management - create, list, restore checkpoints. Invoke when starting risky changes or needing safe recovery points.
Initializes a public GitHub repo with CITATION.cff, citation.bib, LICENSE (MIT), AMA+APA CSL files, README.md with badges and citation block, and gh repo description. Use when creating a new repo, setting up project metadata, or when user mentions "repo init", "badges", "citation", "CITATION.cff", or "license setup".
Enforces a three-phase RPI workflow (Research, Plan, Implement) for code changes. Use when tackling non-trivial features, refactors, or bug fixes in medium-to-large codebases to prevent context waste and ensure correctness over speed.
Unified SuperClaude workflow skill with subcommands for analysis, building, documentation, git, implementation, and more. Use for any structured development workflow.
Use when deploying Docker services on the local VM (hostname: vm, Pop!_OS) with Traefik reverse proxy and Homepage dashboard. Covers crane image workflow, Traefik file-provider registration, Homepage services.yaml entries, and compose templates on the traefik-proxy network.