Skills
98,067 packaged capabilities and playbooks for agents. Install any of them with one command.
Development
Expert knowledge for Drizzle ORM - the lightweight, type-safe SQL ORM for edge and serverlessUse when "drizzle, drizzle orm, drizzle-kit, drizzle schema, drizzle migration, drizzle relations, sql orm typescript, edge database, d1 database, orm, database, typescript, sql, edge, serverless, d1, postgres, mysql, sqlite" mentioned.
Guide for performing Node.js version upgrades in the Electron project. Use when working on the roller/node/main branch to fix patch conflicts during `e sync --3`. Covers the patch application workflow, conflict resolution, analyzing upstream Node.js changes, building, running the Node.js test suite, and proper commit formatting for patch fixes.
Enterprise-grade systems with microservices, Kubernetes, Terraform, and AI Native methodology. For multi-feature initiatives spanning a release timeline, combine with /sprint master-plan (v2.1.13) to group features into a single 8-phase sprint container with shared scope/budget and 4 auto-pause triggers (QUALITY_GATE_FAIL / ITERATION_EXHAUSTED / BUDGET_EXCEEDED / PHASE_TIMEOUT). Triggers: microservices, k8s, terraform, monorepo, AI native
Helps understand and write EAS workflow YAML files for Expo projects. Use this skill when the user asks about CI/CD or workflows in an Expo or EAS context, mentions .eas/workflows/, or wants help with EAS build pipelines or deployment automation.
Generates production-ready FastGPT workflow JSON from natural language requirements. Uses AI-powered semantic template matching from built-in workflows (document translation, sales training, resume screening, financial news). Performs three-layer validation (format, connections, logic completeness). Supports incremental modifications to add/remove/modify nodes. Activates when user asks to "create FastGPT workflow", "generate workflow JSON", "design FastGPT application", or mentions workflow automation, multi-agent systems, or FastGPT templates.
Sync and audit design-system contracts between Figma and code. Use this skill when the user asks about Figma Variables, styles, tokens, component libraries, variants, Code Connect, effect styles, naming drift, token exports/imports, library governance, or keeping implementation and Figma design systems aligned under the UI/UX skill system.
A complete reference for FiveM native functions, auto-updated from [cfxnatives.dev](https://cfxnatives.dev) and organized by namespace. Use this skill to get accurate native signatures, parameters, return types, and examples when developing FiveM resources.
Upload images to img402.dev for embedding in GitHub PRs, issues, and comments. Images 1MB or under are free AND permanent — no payment, no auth — which covers virtually every screenshot. Larger images (up to 10MB) are free for 30 days, or permanent for $0.01 USDC via x402. Use when the agent needs to share an image in a GitHub context — screenshots, mockups, diagrams, or any visual. Triggers: "screenshot this", "attach an image", "add a screenshot to the PR", "upload this mockup", or any task producing an image for GitHub.
Query trending GitHub projects, with filtering by programming language, date range, and spoken language (documentation language). Use this skill when the user wants to know about GitHub trending projects today/this week/this month, trending repositories in a specific language, or trending projects with Chinese documentation.
Godot 4.x game architecture, authoritative state, data ownership, managers, autoloads, scene flow, and save/load boundaries. Use when deciding where runtime state lives, splitting scene-local and persistent systems, coordinating menus and gameplay scenes, designing new-run/reset behavior, or implementing versioned persistence.
Write or improve a README.md for an open source project. Use this for any request to create, rewrite, review, or polish a README, project description, or "getting started" docs for a library, framework, or tool, or to make an open source project more popular or attractive to users. Do NOT use for internal/private repo docs, API reference generation, or CHANGELOG/CONTRIBUTING files.
Quantizes LLMs to 4-bit (also 3-bit) with GPTQ using group-wise quantization (group size 128 by default), via AutoGPTQ and transformers. Covers loading pre-quantized GPTQ models, quantizing your own model, choosing group size, selecting ExLlamaV2, Marlin, or Triton kernels, and QLoRA fine-tuning with PEFT. Use when fitting 70B-class models onto limited or consumer GPUs, cutting memory about 4x versus FP16, speeding up inference, finding pre-quantized checkpoints on HuggingFace, or fine-tuning a quantized model with LoRA. For slightly better accuracy on newer GPUs use AWQ instead, and for simple 8-bit or on-the-fly quantization use bitsandbytes.
Automatically detect GPU vendor, find appropriate PyTorch container image, launch with correct mounts, and validate GPU functionality. Supports NVIDIA, Ascend, Metax, Iluvatar, and AMD/ROCm. Use when user says "setup container", "start pytorch container", or invokes /gpu-container-setup.
Explain an unfamiliar codebase through a repository briefing, dataflow trace, or change-history reconstruction. Use when asked for a repo tour, how something works, where a value comes from, or why and how a feature or change was introduced. Do NOT use for repository health (use repo-eval) or diff review.
Guides GRPO (Group Relative Policy Optimization) fine-tuning of language models with the TRL library, including GRPOTrainer configuration, composing multiple reward functions (correctness, format, length, style), dataset prep in chat format, Unsloth setup, LoRA merging, and monitoring reward, reward_std and KL. Use when training a model to follow a strict output format such as XML or JSON. Use when teaching verifiable tasks like math or code with objective correctness rewards. Use when improving chain-of-thought reasoning with custom reward functions. Use when debugging flat rewards, mode collapse, or OOM in GRPO runs. Do not use for plain supervised fine-tuning (use SFT) or when high-quality preference pairs exist (use DPO or PPO).
Official GSAP skill for GSAP plugins — registration, ScrollToPlugin, ScrollSmoother, Flip, Draggable, Inertia, Observer, SplitText, ScrambleText, SVG and physics plugins, CustomEase, EasePack, CustomWiggle, CustomBounce, GSDevTools. Use when the user asks about a GSAP plugin, scroll-to, flip animations, draggable, SVG drawing, or plugin registration.
Official GSAP skill for timelines — gsap.timeline(), position parameter, nesting, playback. Use when sequencing animations, choreographing keyframes, or when the user asks about animation sequencing, timelines, or animation order (in GSAP or when recommending a library that supports timelines).
Constrains LLM output during generation with Guidance (Microsoft Research), using regex, select() choices, context-free grammars, token healing, and @guidance functions, with Anthropic, OpenAI, Transformers, and llama.cpp backends. Use when you need generated text to match a regex or fixed format such as dates, emails, or IDs. Use when you need guaranteed valid JSON, XML, or code from a model. Use when building multi-step generation workflows or ReAct-style agents with Python control flow. Use when classifying text into fixed categories with select(). Use when running local models and wanting grammar constraints. Not for Pydantic validation with automatic retries; use Instructor for that.
Use alongside the `hardhat` skill when the project depends on `@nomicfoundation/hardhat-toolbox-viem`. Covers the viem clients exposed on `network.create()`, contract interaction (`viem.deployContract`, `read`, `write`, `getContractAt`), and `viem.assertions` (revert / event / balance assertions).
Use when planning tests for Arduino, ESP32, ESP32-S3, Pico, or Pico W firmware; deciding what can run off-target; using Wokwi or simulation; writing host fakes; or accepting a physical hardware feature without confusing compile proof with system proof.
Half-Quadratic Quantization for LLMs without calibration data. Use when quantizing models to 4/3/2-bit precision without needing calibration datasets, for fast quantization workflows, or when deploying with vLLM or HuggingFace Transformers
Wraps existing PyTorch training scripts with HuggingFace Accelerate (Accelerator class, accelerate config, accelerate launch) so the same code runs on CPU, single GPU, multi-GPU, multi-node, TPU, or Apple MPS. Covers device placement, FP16/BF16/FP8 mixed precision, gradient accumulation, distributed checkpointing, and switching between DDP, DeepSpeed ZeRO, FSDP, and Megatron backends. Use when converting a single-GPU script to multi-GPU, enabling mixed precision, configuring DeepSpeed ZeRO or FSDP, or setting up gradient accumulation. Use when one script must run on different hardware. Not for callback-heavy training loops (use PyTorch Lightning) or multi-node orchestration with hyperparameter tuning (use Ray Train).
Provides the HuggingFace Tokenizers library (Rust core with Python and Node.js bindings) for training and using BPE, WordPiece, and Unigram tokenizers. Covers the pipeline of normalizers, pre-tokenizers, models, post-processors and decoders, padding and truncation, batch encoding, alignment tracking, and conversion to transformers PreTrainedTokenizerFast. Use when training a custom tokenizer or vocabulary on a new corpus, tokenizing large text corpora quickly, mapping tokens back to character offsets for NER or question answering, or configuring normalization and special tokens. Use when wrapping a custom tokenizer for transformers. For SentencePiece models or tiktoken, use those tools instead; for loading a pretrained tokenizer only, AutoTokenizer is enough.
Implement, tune, debug, or validate soft wall-aware fog of war and gameplay perception in Three.js action games. Use for orthographic or isometric visibility masks, obstacle-aware line of sight, player and enemy vision ranges, hidden-enemy targeting rules, fog shader artifacts such as spokes or seams, mobile ray budgets, lifecycle and menu-state integration, and deterministic fog-of-war tests.
Use when the user wants several open issues built by agents and turned into pull requests — "work through the issues", "implement the open issues", "clear the backlog", "open PRs for #12 and #15" — optionally at an effort level ("/implement-issues low", "at medium effort"), and when plan pull requests it opened are approved or have comments to answer. Works in Claude Code, Codex, or any agent that can dispatch subagents. Not for one issue the user is building themselves in this session.
Implements and trains LLMs using Lightning AI's LitGPT with 20+ pretrained architectures (Llama, Gemma, Phi, Qwen, Mistral). Use when need clean model implementations, educational understanding of architectures, or production fine-tuning with LoRA/QLoRA. Single-file implementations, no abstraction layers.
Extracts structured, validated data from LLM responses using the Instructor Python library with Pydantic response models, including nested models, enums, custom validators, automatic retries with validation error feedback, and streaming of partial objects or iterables. Works with Anthropic, OpenAI, and local Ollama models. Use when pulling typed fields or entities out of free text, when classifying text into fixed categories, when an LLM must return JSON that passes schema validation, when failed extractions need automatic retry, or when streaming partial structured results. Not for prompt optimization (use DSPy) or building multi-step chains (use LangChain).
ioredis v5 reference for Node.js Redis client — connection setup, RedisOptions, pipelines, transactions, Pub/Sub, Lua scripting, Cluster, and Sentinel. Use when: (1) creating or configuring Redis connections (standalone, cluster, sentinel), (2) writing Redis commands with ioredis (get/set, pipelines, multi/exec), (3) setting up Pub/Sub or Streams, (4) configuring retryStrategy, TLS, or auto-pipelining, (5) working with Redis Cluster options (scaleReads, NAT mapping), or (6) debugging ioredis connection issues. Important: use named import `import { Redis } from 'ioredis'` for correct TypeScript types with NodeNext.
Expert Java developer for implementing LeetCode problems, data structures, and algorithms. Helps write clean, efficient Java code following best practices for competitive programming and interview preparation. Use when writing or implementing Java solutions.
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.