持續整合/部署

Design architecturally optimal ClickHouse data warehouses. Use when creating schemas, modeling data, optimizing queries, or making architectural decisions about ClickHouse tables, materialized views, projections, or dictionaries.

Brief description with trigger keywords. Use when [specific situations].

Optimize educational content specifically for Google NotebookLM processing, book generation, quiz creation, and flashcard development with enhanced metadata, cross-references, and AI-friendly formatting.

Use when facing 3+ independent failures that can be investigated without shared state or dependencies - dispatches multiple Claude agents to investigate and fix independent problems concurrently

Write or upgrade conventional Python docstrings for public modules, classes, functions, methods, and properties. Docstrings must be user-facing, mkdocstrings-friendly, include runnable examples, and must not repeat type hints.

End-to-end software development workflow agent using memex-cli with multiple AI backends. Orchestrates 5 stages: (1) Requirements Analysis (Claude), (2) Functional Design (Claude), (3) UX Design (Gemini), (4) Development Planning (Codex), (5) Implementation & QA (Codex). Use for complete project development from idea to deployment.

Manages Vue state with Pinia including stores, getters, actions, and plugins. Use when building Vue applications needing centralized state, sharing state between components, or replacing Vuex.

Use when setting up git workflows for Claude Code - installs pre-commit hooks, commit templates with Claude attribution, PR templates, branch naming helpers, and Claude-specific gitignore entries

This skill should be used when orchestrating tmux sessions for multi-component development workflows. Use when creating complex multi-window, multi-pane tmux layouts via the Bash tool, especially for projects requiring simultaneous backend/frontend development, microservices, or parallel task execution. This skill provides patterns for session creation, pane management, and command orchestration using the user's custom tmux configuration.

Node.js backend development with TypeScript, Express/Fastify servers, routing, middleware, and database integration. Use when building REST APIs with TypeScript, creating Express/Fastify servers, or implementing microservices architectures.

Comprehensive purple team security assessment skill that analyzes codebases like a skilled penetration tester, identifies attack vectors, and provides remediation guidance. Use this skill when: (1) Performing security audits or code reviews, (2) Identifying vulnerabilities before deployment, (3) Running purple team exercises (offensive analysis + defensive remediation), (4) Checking for OWASP Top 10 vulnerabilities, (5) Scanning for hardcoded secrets, API keys, or credentials, (6) Analyzing authentication/authorization flaws, (7) Reviewing cryptographic implementations, (8) Auditing infrastructure-as-code (Terraform, K8s, Docker), (9) Generating security reports in Markdown, HTML, or JSON format. Triggers: "security audit", "purple team", "penetration test", "find vulnerabilities", "security review", "attack vectors", "remediate vulnerabilities", "OWASP", "security scan".

Sets up a CI/CD pipeline for Python projects using GitHub Actions. Includes templates for PR checks, deployment via SSH, pre-commit hooks, and email notifications. Use this skill when a user wants to create a new CI/CD pipeline for a Python project.

Weekly meal prep and shopping assistant that selects 3 recipes with varied proteins, generates categorized shopping lists, and creates meal prep plans distinguishing Sunday prep from weeknight cooking. Use when planning weekly meals or creating shopping lists.

Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup. Presents exactly 4 options: merge locally, create PR, keep as-is, or discard. Do NOT use when tests are failing or work is incomplete - fix issues first before finishing the branch.

Generate the full production Dockerfile for the backend with multi-stage build, proper dependencies, and Hugging Face Spaces compatibility.

Audits code against CI/CD style rules, quality guidelines, and best practices, then rewrites code to meet standards without breaking functionality. Use this skill after functionality validation to ensure code is not just correct but also maintainable, readable, and production-ready. The skill applies linting rules, enforces naming conventions, improves code organization, and refactors for clarity while preserving all behavioral correctness verified by functionality audits.

GitLab CI/CD pipeline patterns including testing stages, Docker builds, caching strategies, secrets management, deployment patterns (blue-green, canary, rolling), and rollback procedures. Use when creating .gitlab-ci.yml, optimizing CI performance, setting up deployment pipelines, or troubleshooting CI failures.

Build MCP servers for integrating external APIs with AI agents

This skill should be used when the user asks about "workflow XML", "planner", "agent dependencies", "workflow parsing", "task orchestration", or needs to understand the workflow system in XSky.

Comprehensive system design skill for creating professional software architecture specifications. Use this skill when asked to design systems (e.g., "Design a chat application", "Design an e-commerce platform", "Create system architecture for X"). Generates complete technical specifications with architecture diagrams, database schemas, API designs, scalability plans, security considerations, and deployment strategies. Creates organized spec folders with all documentation following professional software engineering standards, from high-level overview down to detailed implementation specifications.