Claude Skills

This skill should be used when the user asks to 'audit for JSR', 'check JSR readiness', 'review JSR config', 'verify package for JSR', 'publish to JSR', 'prepare for JSR publishing', 'JSR compliance check', 'run JSR audit', or wants to ensure their Deno/TypeScript package meets JSR standards before publishing.

Run code on cloud GPUs with GPU CLI. Use when user wants to run Python/ML code on remote GPUs, install GPU CLI, configure gpu.toml, troubleshoot pods, or manage cloud GPU resources via RunPod.

Use for any code implementation task. Enforces RED-GREEN-REFACTOR cycle. "If you didn't watch it fail, you don't know if it tests the right thing."

Use when interacting with Hetzner Cloud via command line - managing servers, networks, volumes, load balancers, firewalls, DNS, or any cloud infrastructure operations with hcloud CLI

Capture moment of clarity before moving forward

Epic design document template. Use when creating new epic documents or need epic structure reference.

This skill should be used when the user asks to "generate image", "create image with nanobanana", "write image prompt", "make icon/pattern/sticker", or "image generation". Crafts effective prompts for nanobanana MCP with Gemini Imagen.

This skill should be used when the user asks to "analyze video", "summarize video", "extract video transcript", "understand video content", "video to text", "describe video", "ask questions about video", or "what happens in this video". Analyzes videos using Google Gemini API with local file upload or YouTube URL input.

Use when writing or reviewing tests - covers test philosophy, condition-based waiting, mocking strategy, and test isolation

Activates when the user asks about Agent Skills, wants to find reusable AI capabilities, needs to install skills, or mentions skills for Claude. Use for discovering, retrieving, and installing skills.

Use when writing or refactoring code, before creating files - enforces separation of pure business logic (Functional Core) from side effects (Imperative Shell) using FCIS pattern with mandatory file classification

Use when writing Go code to interact with Gitea API - automation, bots, integrations, migrations, or programmatic git forge operations

This skill MUST be invoked before any completion claim, success report, or task transition. Mandates 5-step evidence protocol. Use when about to claim "done", "complete", "working", "fixed", or any success state.

This skill should be used when the user asks to "learn domain concepts", "domain vocab", "learn [domain] vocabulary", "what are key concepts in [domain]", "explore domain", "divergent exploration", or wants to build conceptual vocabulary for AI prompting. Provides structured learning of domain terminology with token priming, divergent exploration, and depth progression.

Prisma relation modeling patterns

Use when creating implementation plans. Mandates exact file paths, complete code samples, and expected output. No vague language allowed.

Setup VSCode for F5 Python debugging in UE5 editor. Use when users need to (1) configure VSCode for UE5 Python debugging, (2) setup debugpy remote debugging workflow, (3) enable breakpoint debugging in UE5 Python scripts, or (4) create VSCode launch/task configurations for UE5.

Explore cutting-edge research in a domain. Use when user says "frontier [domain]", "latest research in [domain]", "cutting edge", "latest in [domain]", "what's new in [field]", "[domain] trends", or wants L4-level current research.