機器學習

Autonomously run quality checks (RuboCop, RSpec) to establish baseline or detect regressions. Triggers before starting development work, after completing changes, when checking code quality, or when explicitly requested. Compares current metrics against previous baseline to identify improvements or regressions in code quality and test coverage.

Create new Claude Code Skills with proper YAML frontmatter, progressive disclosure structure, and complete directory organization. Use when you need to build custom skills for specific workflows, generate skill templates, or understand the Claude Skills specification.

Expert knowledge for modern browser technologies including HTML5, CSS (Grid, Flexbox, Container Queries, Scroll-Driven Animations, Anchor Positioning), Progressive Web Apps (PWA, Service Workers, Web Push, File System Access), and Browser APIs (Canvas, Geolocation, Web Storage, Web Share, Drag and Drop). Use when working with web standards, client-side features, responsive design, offline capabilities, or native-like web experiences.

HtmlGraph session tracking and work continuity for Codex. Ensures proper activity attribution, feature awareness, and continuous work tracking across sessions.

QA specialist agent for test planning, execution, and regression analysis

This is a test fixture with intentional violations for regression testing. It contains <angle brackets> which trigger B5, and has an unexpected frontmatter property which triggers W2.

This skill should be used when the user asks to "generate gitlab-ci.yml", "create GitLab CI pipeline", "configure GitLab CI/CD", "use To-Be-Continuous templates", "setup TBC templates", "create CI/CD for Python/Node/Go/Java project", "configure Docker build in GitLab", "setup Kubernetes deployment in GitLab", "add SonarQube to GitLab CI", "configure Terraform with GitLab", or mentions "TBC", "To-Be-Continuous", "Kicker".

Build desktop and mobile apps with Tauri v2, a Rust-based framework using web frontends (HTML/CSS/JS). Use when creating new Tauri projects, adding Rust commands, configuring IPC between frontend and backend, managing plugins, or building/distributing apps. Triggers for file patterns like src-tauri/*, tauri.conf.json, or questions about Tauri commands, events, capabilities, or plugins.

Defines context handover format when workers hit turn limit. Posts structured handover to GitHub issue comments enabling replacement workers to continue seamlessly.

Use when creating Docusaurus plugins (remark, rehype, theme, content, lifecycle) to extend markdown, modify HTML, or add custom functionality

Deploy F# applications with Tailscale sidecar for private network access. Use when: "deploy", "production", "Tailscale", "docker-compose", "Docker", "Portainer", "private network", "deployment", "ship it", "go live", "home server", "self-host", "container", "release". Creates docker-compose.yml with app + Tailscale sidecar for secure access. No public ports, no authentication needed - Tailscale handles it.

Generate XML documentation comments for C# code based on signatures, naming conventions, and code structure. Use when user requests documentation or for public APIs. (project, gitignored)

Generate application wiring: entry points, dependency injection, infrastructureconfiguration. Creates main.py, container, Dockerfile, docker-compose.yaml.Topology-aware (monolith/microservices/library). Use when: finalizing generation,creating deployable system.

Generate professional PDF/HTML reports with charts, tables, and narrative from data. Supports templates, branding, and automated report generation.

Set up Sveltia CMS - the lightweight Git-backed CMS successor to Decap/Netlify CMS with 5x smaller bundle (300 KB), GraphQL performance, and 260+ fixed predecessor issues. Framework-agnostic for Hugo, Jekyll, 11ty, Gatsby, Astro. Use when: adding CMS to static sites, migrating from Decap/Netlify CMS, enabling non-technical editors, or troubleshooting OAuth authentication failures, YAML parse errors, CORS/COOP policy problems, content not listing issues.

Synthesize all quality assessments (risk, test-design, traceability, NFR) into evidence-based gate decision (PASS/CONCERNS/FAIL/WAIVED) with comprehensive rationale. Generates both YAML (CI/CD) and Markdown (human review) reports with action items. Use during final quality review to make go/no-go deployment decisions based on comprehensive quality evidence.

Create and edit draw.io diagrams with real-time browser preview using @next-ai-drawio/mcp-server. Trigger when user asks to: - Create flowcharts, process diagrams, or workflow visualizations - Generate system architecture diagrams or component diagrams - Visualize code structure, microservices, or deployment architecture - Edit or modify existing draw.io diagrams - Export diagrams to .drawio files Keywords: diagram, flowchart, architecture, draw.io, drawio, visualization, UML, sequence diagram, component diagram

Use SAP-RPT-1-OSS open source tabular foundation model for predictive analytics on SAP business data. Handles classification and regression tasks including customer churn prediction, delivery delay forecasting, payment default risk, demand planning, and financial anomaly detection. Use when asked to predict, forecast, classify, or analyze patterns in SAP tabular data exports (CSV/DataFrame). Runs locally via Hugging Face model.

Create machine-readable SDS YAML that maps cleanly to SEA DSL.

Python/pytest TDD specialist for test-driven development workflows. Use when writing tests, auditing test quality, running pytest, or generating test reports. Integrates with uv and pyproject.toml configuration.