持續整合/部署

Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor/pipeline/sequence/context/expert parallelism. Production-ready framework used for Nemotron, LLaMA, DeepSeek.

Mobile game development principles. Touch input, battery, performance, app stores.

Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training

Expert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, sparse attention

Pragmatic coding standards - concise, direct, no over-engineering, no unnecessary comments

Use when writing or changing tests, adding mocks, or tempted to add test-only methods to production code - prevents testing mock behavior, production pollution with test-only methods, and mocking without understanding dependencies

Web application testing principles. E2E, Playwright, deep audit strategies.

Production deployment principles and decision-making. Safe deployment workflows, rollback strategies, and verification. Teaches thinking, not scripts.

Architectural reasoning framework: requirements analysis, trade-off evaluation, and ADR-based decision documentation. NOT a pattern catalog, but a thinking framework. Use when designing system architecture, choosing patterns, or making structural decisions.

Use when starting any conversation - establishes mandatory workflows for finding and using skills, including using Skill tool before announcing usage, following brainstorming before coding, and creating TodoWrite todos for checklists

Renaming a C file or associated header (.h) file requires additional steps in decompilation projects or the build will fail. Use this skill when you want to rename a C file and/or a header file.

Captures session learnings, decisions, and context to a markdown file for future agent ramp-up. Use when user says "ppage", "page context", "save context", "capture learnings", or before ending a substantial work session.

Multi-specialist collaborative analysis for complex decisions. Spawns parallel subagents with diverse domain expertise to analyze a topic from multiple angles. Use when user says "multi-mind", or for complex architecture decisions, technology choices, strategic planning, or any multi-faceted problem with no obvious right answer.

This skill should be used when the user asks to "create a justfile", "write just recipes", "configure just settings", "add just modules", "use just attributes", "set up task automation", mentions justfile, just command runner, or task automation with just.

This skill should be used when the user asks to "update dependencies", "update npm packages", "run taze", "upgrade node packages", "check for outdated packages", "update package.json", or mentions dependency updates, npm/pnpm/yarn package upgrades, or taze CLI usage.

Generate web assets including favicons, app icons (PWA), and social media meta images (Open Graph) for Facebook, Twitter, WhatsApp, and LinkedIn. Use when users need icons, favicons, social sharing images, or Open Graph images from logos or text slogans. Handles image resizing, text-to-image generation, and provides proper HTML meta tags.

This skill should be used when the user asks to "refactor duplicate code", "apply DRY principles", "eliminate code repetition", "extract common functionality", or mentions code duplication, similar patterns, repeated logic, or reusable abstractions.

This skill should be used when the user asks to "configure TypeScript", "fix type errors", "use dayjs", "add type definitions", "set up React with TypeScript", mentions ".ts" or ".tsx" files, or asks about TypeScript best practices or TypeScript-specific tooling.

Creates and manages ephemeral Neon databases for testing, CI/CD pipelines, and isolated development environments. Use when building temporary databases for automated tests or rapid prototyping.

Creates a fully functional Drizzle ORM setup with a provisioned Neon database. Installs dependencies, provisions database credentials, configures connections, generates schemas, and runs migrations. Results in working code that can immediately connect to and query the database. Use when creating new projects with Drizzle, adding ORM to existing applications, or modifying database schemas.