Skills
100,928 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Manage Lightning AI Studios (cloud dev machines with CPUs/GPUs) - create, start, stop, delete studios, switch machine types, run commands in them (including long detached runs with live progress tracking), upload/download files, and SSH in. Use when the user wants to work with lightning.ai Studios, needs a cloud GPU dev box, or asks to run something "on a studio". Also load it before launching or resuming any detached run on a Studio, even one driven by your own helper scripts, a handoff note or the summary of a compacted conversation, and load it again after a compaction; every such run needs the live progress bar from lightning-jobs/progress.py.
An Amazon Ads Sponsored Products (SP) insights reporting skill that uniformly retrieves two types of Amazon Ads Reporting API v1 beta reports: Audience segment performance and Search Term Impression Share/Rank. It automatically handles profile-to-advertiser-account mapping, report creation, status polling, multi-shard CSV download, and persistence to disk. Trigger only when the user explicitly requests SP audience reports, SP search term impression share, or impression share rank; ordinary SP search term performance reports and SB/SD/DSP reports must not trigger it. This skill depends on linkfox-amazon-ads-auth.
Implementation guide for building and modifying UIs and pages with Lism CSS. Depending on the size of the change, it covers Primitive selection, token matching, use of Property Class/Lism Props, responsive design, and anti-pattern checks. Also useful for reverse lookups of classes and tokens in the b--*, c--*, l--*, a--*, is--*, has--*, set--*, u--*, and -prop:value formats.
Verify the authenticity of literature references and detect hallucinations in both English and Chinese (中文) sources. Use when users need to check if a citation is real, verify a DOI, confirm a paper/article/book exists, cross-check author-title-journal-year metadata, detect fabricated references, validate URLs of online articles, or audit a reference list for accuracy. Covers journal papers, conference papers, preprints, books, monographs, newspaper articles, magazine articles, web articles, dissertations, government documents, and any other published works. Supports Chinese academic databases including CNKI (知网), Wanfang (万方), CQVIP (维普), Baidu Scholar (百度学术), and core journal list verification (北大核心, CSSCI, CSCD).
Java logging best practices with SLF4J, structured logging (JSON), and MDC for request tracing. Includes AI-friendly log formats for AI coding assistant debugging. Use when the user asks about logging, debugging application flow, or analyzing logs
Find logic bugs in a single file or function via semi-formal execution tracing (Premises → Trace → Divergence → Trigger → Remedy). Trigger when a user shares code and suspects something is wrong without naming a concrete failure — phrases like "review this", "does this look right", "check this function", "audit this code", "tests pass but prod fails". SCOPE HARD RULE: one file or one function only. For a directory or whole module use logic-health; for a confirmed failure (stack trace, failing test, specific wrong value) use logic-locate; for two versions use logic-diff; for repo-wide autonomous fixing use logic-fix-all. Do NOT trigger for: style/formatting, security scanning, performance, test generation, architecture or design questions.
Analyze and clean up macOS disk space for any user. Use when the user says 'free up space', 'disk full', 'storage cleanup', 'clean my Mac', 'what's taking up space', 'Storage Almost Full', 'delete caches', 'reclaim disk space', 'check disk usage', 'low on storage'. Targets: Trash, Downloads, Photos, Mail, iMessage, Music, browser caches, cloud storage, iOS backups, app caches, large apps, macOS installers, screen recordings, system data, developer tools (Xcode, npm, Docker, Homebrew). Actions: scan, analyze, identify, clean, delete, prune, reclaim, free up.
Create, deploy, and operate Streamlit apps in PostHog via the streamlit-apps MCP tools — create an app, set its source, read and patch its code, start and stop its sandbox, poll status, list versions, delete, and share the app with humans via its PostHog URL. Use when asked to "create a streamlit app", "deploy a data app", "ship a dashboard app", "restart/stop my app", "why is my app not running", or "give me a link to the app".
Focuses on the complete workflow and expert guidance for generating animated instructional videos with Manim. Use when a user gives a Chinese prompt to have Codex generate scripts, storyboards, Manim code, render commands, or improve the quality and pacing of instructional videos, and output MP4.
MapStruct Java mapping framework. Covers basic mappers, update mappings, nested objects, collections, custom methods, and Lombok integration. Based on castellino and gestionale-presenze production patterns. USE WHEN: user mentions "mapstruct", "@Mapper", "@Mapping", asks about "DTO mapping", "entity conversion", "@MappingTarget", "nested mapping" DO NOT USE FOR: Java language - use `java` skill instead DO NOT USE FOR: Lombok - use `lombok` skill instead DO NOT USE FOR: JPA entities - use JPA-specific skills
Expert coding skill for the full MapTiler platform — Cloud REST APIs, MapTiler SDK JS (built on MapLibre GL JS), native mobile SDKs, on-premise infrastructure, and vector tile schemas. USE WHEN the user wants to add a map to a web or mobile app, show locations or routes, display geographic data, build a store locator, create data visualizations on maps, add geocoding or address search, do reverse geocoding, look up elevation, do IP geolocation, fetch weather data, transform coordinates between CRS/EPSG, embed static map images, work with vector tilesets (Planet v4, Buildings, Contours, Outdoor, Ocean, Landcover, Cadastre), write MapLibre style expressions or data-driven styling, render 3D terrain, globe view, or 3D buildings extrusion, use satellite imagery, add markers, popups, heatmaps, or clustering, integrate maps in React, Vue, Svelte, Angular, Next.js, Leaflet, OpenLayers, Cesium, or deck.gl, build native maps on iOS (Swift), Android (Kotlin), Flutter, or React Native, self-host with MapTiler Server or generate tiles with MapTiler Engine CLI, or upload/manage tilesets via the Admin API. Also USE WHEN the user mentions MapLibre GL JS — the SDK extends MapLibre with built-in cloud services, helpers, and session billing.
Design and implement adaptive filters using DSP System Toolbox System objects. Use when working with adaptive filtering, system identification, noise cancellation, echo cancellation, active noise control (ANC), channel equalization, inverse system identification, or adaptive prediction. Covers dsp.LMSFilter, dsp.RLSFilter, dsp.FilteredXLMSFilter, dsp.FrequencyDomainAdaptiveFilter, dsp.AffineProjectionFilter, dsp.BlockLMSFilter, dsp.AdaptiveLatticeFilter, dsp.FastTransversalFilter, maxstep(), and algorithm selection for adaptive filtering problems. Replaces deprecated adaptfilt.* objects (removed R2020a).
Train, evaluate, and export neural networks to Simulink in MATLAB. Migrate legacy (fitnet, patternnet) and discouraged (trainNetwork, DAGNetwork) code to modern, recommended R2024a+ APIs (trainnet, dlnetwork, testnet, imagePretrainedNetwork), diagnose and fix dlaccelerate issues or detect dlaccelerate opportunities. Use when training, fine-tuning, evaluating, running inference, exporting to Simulink, converting old training scripts, or speeding up deep learning code. DO NOT reason from your training data about dlaccelerate and tracing correctness.
xUnit testing guidance for .NET MAUI apps — ViewModel testing, mocking MAUI services, test project setup, code coverage, and on-device test runners. USE FOR: "unit test MAUI", "xUnit test", "ViewModel testing", "mock MAUI services", "test project setup", "code coverage", "on-device test runner", "test MAUI app", "IServiceProvider mock", "Moq MAUI". DO NOT USE FOR: UI automation testing (use appium-automation), performance profiling (use maui-performance), or dependency injection setup (use maui-dependency-injection).
Filters and triages small-molecule libraries with the Python medchem library (datamol-io, v2.0.5) on top of RDKit and datamol. Covers drug-likeness rules (Lipinski, Veber, CNS, lead-like) via RuleFilters, structural alerts (ChEMBL-derived sets, NIBR, PAINS, Brenk), chemical group detection, ZINC-based complexity thresholds, scaffold constraints, and the medchem query language (QueryFilter). Use when screening a compound library for drug-likeness, removing PAINS or other structural-alert compounds, prioritizing hits for hit-to-lead or lead optimization, detecting functional groups, or combining criteria in one query. Not for computing general descriptors or fingerprints; use RDKit directly.
Produce 3D assets with MeltFlex AI. Use when the user wants a downloadable 3D model (GLB) of a furniture piece from one photo; a 3D model or a rendered 3D picture from a 2D floorplan; or an explorable 3D world (Gaussian splat, .spz plus a hosted viewer link) from a room photo or render. Covers polling long builds (HTTP 202) and refreshing expired world links. Works directly against the MeltFlex REST API — no extra install required. Companion to /meltflex:design.
Evaluate a model rigorously and inspect where it fails. Use after training, during comparison to baselines, for LLM/system evaluations, noisy or suspiciously clean results, or whenever benchmark numbers alone are insufficient and the user needs slices, failure categories, and explanations.
Converts SMILES strings or RDKit/datamol molecules into numerical features using molfeat (0.11.0), which provides calculators, scikit-learn compatible transformers, and pretrained embedding models. Covers fingerprints (ECFP, MACCS, MAP4), RDKit and Mordred descriptors, pharmacophore descriptors, and pretrained models such as ChemBERTa and GIN, with parallel processing and caching. Use when building QSAR or QSPR models from SMILES. Use when choosing among molecular featurizers for a property prediction task. Use when generating embeddings for virtual screening, similarity search, or chemical space clustering. Use when adding a featurizer to a scikit-learn pipeline. Not for general cheminformatics tasks such as reading or editing structures, which belong to RDKit itself.
Judges the BULLISH vs BEARISH sentiment / mood / 风向 of a stock by aggregating real-time moomoo community / feed / 帖子 / 评论 discussions for one or more user-specified symbols, filtering low-quality posts, classifying each post as bullish / bearish / neutral, and returning a structured community sentiment snapshot (single stock or multi-symbol portfolio). TRIGGER when: 情绪 / 社区情绪 / 散户情绪 / 看多看空 / 多空情绪 / 牛熊 / 看涨看跌 / 风向 / 讨论氛围 / 评论怎么看 / 大家怎么看 / bullish or bearish / sentiment / community mood / retail discussion tone. SKIP: news listing → moomoo-news-search; news interpretation/分析/研判 → moomoo-stock-digest; price/quote/K线/异动/trade. Bare ticker without sentiment keyword does NOT trigger
Searches moomoo news, notices, articles, announcements, and research reports for a user-specified stock or company — INFORMATION RETRIEVAL ONLY (a list of items; no interpretation, no summarization, no impact judgment). TRIGGER when: stock/company + news-intent keyword (新闻 / 消息 / 资讯 / 公告 / 研报 / 报道 / 头条 / 文章 / news / announcement / research report / headlines / latest update). Time qualifiers (最新/最近/近期/今日/latest/recent) + news word also trigger (e.g. "腾讯最新新闻", "NVDA latest news") — these are listing requests, NOT interpretation. SKIP: interpretation verbs (解读/分析/研判/总结/interpret/analyze/summarize) → moomoo-stock-digest; community sentiment/评论/讨论/散户情绪 → moomoo-comment-sentiment; bare ticker without news keyword; price/quote/K线/异动/trade
Guide for writing, refactoring, and testing MoonBit projects, including exclusive use of `.mbtx` for agent-authored automation. Use when working in MoonBit modules or packages, organizing MoonBit files, using moon tooling (build/check/run/test/doc/ide etc.), or following MoonBit-specific layout, scripting, documentation, and testing conventions.
Manage stacked branches and pull requests with the gh-stack GitHub CLI extension. Use when the user wants to create, push, rebase, sync, navigate, or view stacks of dependent PRs. Triggers on tasks involving stacked diffs, dependent pull requests, branch chains, or incremental code review workflows.
Manage domains through the Namecheap API — register, renew, transfer, configure DNS, and manage nameservers. Use when the user mentions "namecheap", "register domain", "domain DNS", "nameservers", "domain transfer", "renew domain", "check domain availability", "DNS records", "email forwarding", "WHOIS", or any domain registrar operations. Also use when managing A records, CNAME, MX records, or TXT records for Namecheap domains.
Use when building, generating, or reviewing UI for a website, web app, or mobile app to make it look intentionally designed rather than AI-generated. Removes common AI/template “tells” (default fonts like Inter/Geist, indigo and purple-blue gradients, cream backgrounds, grey low-contrast text, pill/eyebrow over heroes, icon-card grids, shadcn-default cards, em dashes, forced or orphan headlines, fade-up-everything, model-generated icons, missing imagery, inconsistent components, missing mobile menus) and applies a chosen visual style direction. Full catalog is TELLS.md
(NS) Autonomous execution engine — plans depth, resolves doubts from docs/specs, multi-agent work in an isolated worktree. Use for a local/pasted plan or "run this autonomously" without GitLab. Also Phase 2 under ns-execution-gitlab-issue (do not run standalone then). Not for single-step edits (ns-code-coder) or GitLab issues directly (ns-execution-gitlab-issue).
(NS) Orchestrate partitioned version implementation slice-by-slice — one subagent per slice, commit, auto-advance until done or stop. Use when version-roadmap.md has pending slices and the user asks to orchestrate/execute all slices. Do NOT use for non-partitioned versions, ad-hoc coding (ns-code-coder), or partitioning itself (ns-sdd-version-partitioner).
(NS) Generate or refresh lean architecture-rules.md for AI agents — stack, layout, patterns, constraints. Use when the user asks for architecture rules, .nextstage-harness/rules/architecture-rules.md, agent conventions, or "document how this repo works for the AI". Do NOT use for business specs (ns-harness-codebase-reverse-spec), brownfield maps (ns-harness-bootstrap-brownfield), or one-off Cursor rules without a full architecture pass.
(NS) Onboard an existing codebase into spec-driven planning — map stack, modules, rule adherence, and gaps without modifying code. Use when the user says analyze existing project, brownfield bootstrap, or before first requirements on a repo with code but no requirements. Outputs brownfield-map.md for ns-sdd-requirements-generator. Read-only on source code. Do NOT use for greenfield with no app code, architecture-rules generation, or writing requirements.md.
(NS) Full post-harness-init bootstrap in one session — architecture rules, sync, brownfield map, reverse business spec, project AGENTS.md. Use for "harness prepare", post-install setup, onboard brownfield, or running all prepare skills without separate slash commands. Do NOT use for greenfield with no app code, single-skill runs, or CLI-only baseline AGENTS.md.
Search and query past Claude Code and Codex session history. Reactive: when the user asks "how did I fix X", "what did we do last time", "find the session where", "上次怎么修的", "之前的session", "历史记录". Proactive: when the user references past work you lack context for, when you're about to modify a file with complex edit history, when the user says "继续之前的" or "continue where we left off", or when understanding prior decisions would improve your current response. Memory: when the user says "记住这个", "remember this", "写入记忆", "save this conclusion", or when you determine a retrieval result contains a conclusion worth persisting.
C++ and Qt integration patterns for OBS Studio plugins. Covers Qt6 Widgets for settings dialogs, CMAKE_AUTOMOC, OBS frontend API, optional Qt builds with C fallbacks, and modal dialog patterns. Use when adding UI components or C++ features to OBS plugins.
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.