Skills
98,067 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Reduce LLM size and accelerate inference using pruning techniques like Wanda and SparseGPT. Use when compressing models without retraining, achieving 50% sparsity with minimal accuracy loss, or enabling faster inference on hardware accelerators. Covers unstructured pruning, structured pruning, N:M sparsity, magnitude pruning, and one-shot methods.
Master ES6+ features including async/await, destructuring, spread operators, arrow functions, promises, modules, iterators, generators, and functional programming patterns for writing clean, efficient JavaScript code. Use when refactoring legacy code, implementing modern patterns, or optimizing JavaScript applications
Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.
First-time onboarding for the cheat-on-money skill. Builds a user profile (skills / available time / starting capital / region / on-camera willingness / language) and creates the .money-state.json state file; prerequisite for all other money- sub-skills. Triggers: "搞钱初始化"/"我想搞钱"/"找AI兼职"/"money init"/"搞钱 setup". **When a user wants to find money-making opportunities but .money-state.json does not exist, route here first.**
Prediction market arbitrage & alpha scout. Searches Polymarket and Kalshi for the same event, runs cross-platform arb math (including fees), and ranks opportunities by profitability. Use when asked to "find arb", "scout markets", "find edge", or scan a specific topic across prediction markets.
Use when working with the Moost framework (moost, @moostjs/event-http/cli/wf/ws, @moostjs/swagger, @moostjs/otel, @moostjs/vite, create-moost) to build decorator-driven TypeScript apps for HTTP, CLI, WebSocket, Workflow or custom events. Covers controllers, @Inherit, handler/param/response decorators, DI, interceptors, pipes, auth guards, framework errors, metadata decorators, Wooks composables, adapter authoring, workflow outlets/state, WebSocket rooms/testing, logging, boot/teardown hooks (@MoostInit, app.addInitHook, @MoostDispose, app.dispose(), graceful shutdown, reload leaks), the Vite plugin (dev/build, middleware, SSR/SPA, deployment, SSR externalization splits: AtscriptDbView instanceof, empty table instead of view, dev "Class is not Injectable"), event context/scoping, Swagger/OpenAPI, and OTel tracing/metrics. Not for plain @wooksjs/* apps without Moost decorators (use the wooksjs skill), nor auth/RBAC — @moostjs/arbac is deprecated; use the aooth packages (@aooth/auth-moost, @aooth/arbac-moost) and their skill.
Expand an image-generation request into a detailed MOSS-Image caption or spatial layout, and compile a local JSON draft into a ready-to-paste prompt. Use for prompt enhancement, image composition and text placement; this skill does not generate images.
Track and nag about Microsoft Outlook email and (optionally) Microsoft Teams messages on a Windows machine, without relying on web versions. Use when the user asks to: (1) monitor inbox/mentions and remind them on Telegram/Teams until dismissed, (2) draft short, personable, low-jargon email replies from an existing Outlook thread, (3) surface action items from the last N days (default 7). Works via Outlook Desktop automation (COM) and optionally Microsoft Graph for Teams if configured.
Metasploit Framework usage methodology (one-liner + interactive). Use when exploiting OS-level vulnerabilities (such as EternalBlue/MS17-010), remote database vulnerabilities (such as PostgreSQL/MySQL RCE), or network service vulnerabilities (SMB/RDP/FTP), when a payload needs to be generated, when a reverse shell handler needs to be started, or for post-exploitation operations. MSF has 2000+ exploit modules covering remote exploitation of Windows/Linux operating systems, databases, and network devices. This skill also covers one-liner quick exploitation and interactive_session interactive operation (handler/meterpreter/post-exploitation). This skill should be used in any scenario involving metasploit, msfconsole, meterpreter, OS-level exploits, remote buffer overflows, payload generation, handlers, or post-exploitation.
Nano (XNO) cryptocurrency wallet operations, transaction analysis, and explorer lookups. Use for send/receive, balances, pending funds, address validation, unit conversion, tx/hash/account lookup, explorer links, and Nano block-lattice questions. Prefer xno-mcp first; use xno-skills CLI as fallback. Configured OWS wallets ARE the assistant's own wallets — never claim you cannot receive or hold Nano.
Move design work through Bitwarden's Product and Design Jira workflow — final designs attached to tickets, the 30/60/90 critique cadence tracked in Figma, status transitions on engineering epics and stories, and the one-off engineering story flow.
Screen incoming NDAs and classify them as GREEN (standard), YELLOW (needs review), or RED (significant issues). Use when a new NDA comes in from sales or business development, when assessing NDA risk level, or when deciding whether an NDA needs full counsel review.
GPU-accelerated data curation for LLM training. Supports text/image/video/audio. Features fuzzy deduplication (16× faster), quality filtering (30+ heuristics), semantic deduplication, PII redaction, NSFW detection. Scales across GPUs with RAPIDS. Use for preparing high-quality training datasets, cleaning web data, or deduplicating large corpora.
User-facing NemoClaw guidance for installing, configuring, operating, securing, monitoring, and troubleshooting NemoClaw sandboxes. Use when users ask about NemoClaw quickstarts, OpenClaw and OpenShell relationships, local inference, remote GPU deployment, sandbox lifecycle, network policy, security posture, agent skills, command reference, or issue triage instructions.
WebSocket integration for NestJS using Socket.IO. Covers gateways, rooms, authentication guards, exception filters, and Redis adapter for scaling. USE WHEN: user mentions "NestJS WebSocket", "NestJS gateway", "Socket.IO NestJS", "real-time NestJS", "NestJS chat", "WebSocket authentication NestJS", asks about "how to implement WebSocket in NestJS" DO NOT USE FOR: React WebSocket hooks - use `react-websocket` instead, Spring WebSocket - use `spring-websocket` instead, raw Node.js WebSocket - use `nodejs` skill
Comprehensive biosignal processing toolkit for analyzing physiological data including ECG, EEG, EDA, RSP, PPG, EMG, and EOG signals. Use this skill when processing cardiovascular signals, brain activity, electrodermal responses, respiratory patterns, muscle activity, or eye movements. Applicable for heart rate variability analysis, event-related potentials, complexity measures, autonomic nervous system assessment, psychophysiology research, and multi-modal physiological signal integration.
Chris Voss FBI negotiation framework — tactical empathy, calibrated questions, mirroring, labeling, and the Ackerman model for B2B sales and deal negotiation. Use when: negotiation, tactical empathy, calibrated questions, mirroring, labeling, accusation audit, ackerman, deal negotiation, price negotiation, never split the difference.
When the user wants to apply NLP to supply chain, extract information from documents, analyze supplier communications, classify items, or process unstructured text. Also use when the user mentions "natural language processing," "NLP," "text mining," "document extraction," "supplier sentiment analysis," "product classification from text," "BERT," "transformers for text," or "chatbots for supply chain." For general ML, see ml-supply-chain.
Turns rough notes, bullet points, voice transcripts, or tweet dumps into a polished, publication-ready blog post. Optionally enriches with Tavily research to add supporting data and credibility to claims. Use when asked to write a blog post from notes, turn rough ideas into an article, expand bullet points into a full post, clean up a voice transcript into a blog, or repurpose a tweet thread as an article. Trigger when a user says "write a blog post from this", "turn these notes into a post", "expand this into an article", "make this publishable", "I have rough notes write a blog", or "clean up this transcript".
Use this skill to query your Google NotebookLM notebooks directly from Claude Code for source-grounded, citation-backed answers from Gemini. Browser automation, library management, persistent auth. Drastically reduced hallucinations through document-only responses.
Builds and edits Nuxt apps and layers that use nuxt-unified-ui: un-form / useForm schema forms, un-card, un-table, launchFormPickerDialog / launchChoicePickerDialog / launchDialog, toast helpers, radXxx radashi auto-imports, and the companion unified stack (ufetch, useUFetch, resource-manager, createUnifiedResourceController). Use when package.json depends on nuxt-unified-ui, nuxt.config extends it, or code uses these APIs. Whenever implementation work is finished, run `/nuxt-unified-ui apply`, which applies the skill file by file to uncommitted work (on dev, main, or master) or to the current branch's changed files, falling back to the whole project when there is no Git worktree or the default branch is clean; users can also invoke it directly.
Odoo OWL frontend engineering guidance covering components, templates, reactivity, hooks, props, plugins, registries, debugging, assets, and OWL 1/2/3 migration. Use when working with Odoo frontend components, @odoo/owl, static/src JavaScript/XML templates, signals, proxy, props, t-on, t-ref, t-model, registries, plugins, or OWL migrations.
Configure LangWatch online evaluations and guardrails for production traffic. Use when the user wants to score live traces or threads, monitor production quality, sample incoming traffic, or synchronously block unsafe requests and responses. Do not use for batch experiments.
Build stateful chatbots with OpenAI Assistants API v2 - Code Interpreter, File Search (10k files), Function Calling. ⚠️ Sunset H1 2026; use openai-responses for new projects. Use when: building stateful chatbots, implementing RAG, or troubleshooting "thread has active run", vector store delays, polling timeouts, or file upload errors.
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.