Skills
101,174 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Produce a structured summary of a video, article, tweet thread, or PDF — TLDR, key points with timestamps/locations, notable quotes, and who should read/watch it. Use when the user asks to summarize a link or file, wants a TLDR, or asks "what does this video/article say" / "is this worth watching".
Report Svelte MCP autofixer opportunities when the user corrects a Svelte or SvelteKit mistake the agent made. Use after corrections to .svelte components, .svelte.ts/.svelte.js modules, or SvelteKit features where a reusable check or fix could prevent the mistake. Do not trigger for initial implementation requests, unrelated bugs, or personal style preferences.
Common utilities and features for Syncfusion JavaScript controls. Use this skill when the user needs to implement animations, drag-and-drop, state persistence, RTL support, localization, globalization, security, templates, and advanced features for Syncfusion JavaScript controls.
Use when choosing or changing architecture, a database, an API style, authentication, caching, queues or background jobs, file and module organisation, hosting, or any external service, or when the user asks what to use, where to deploy, how to scale, or whether to split into services
Server-state guidance for queries, mutations, caching, optimistic updates, SSR, persistence, testing, and all official adapters. Use when a project uses TanStack Query or when the task needs its setup, implementation, debugging, testing, migration, or integration guidance.
Build features and bug fixes test-first, with tests that protect behavior. Use when about to implement a feature or bug fix that adds or changes behavior users or callers rely on, including requests that do not mention tests or TDD. Not for removing features or code, behavior-preserving refactors, design or UX exploration, prototypes, styling, copy, configuration that leaves relied-on behavior unchanged, or data cleanup.
Configure and troubleshoot Teleport auto-discovery for AWS EC2 instances, AWS EKS clusters, and Azure VMs via the Teleport discovery Terraform module. Use to set up or extend auto-discovery, add a region, tag, or subscription, apply the discovery Terraform, check enrollment status, or diagnose why cloud resources are not enrolling. Not for GCP, AWS RDS or database discovery, static or manual join-token enrollment, or Teleport configuration unrelated to discovery.
Research public Telegram channels and groups with TGStat. Use when discovering communities by topic, comparing public audience/reach/activity, checking a known @username, or shortlisting ad and outreach sources. The connector supplies TGSTAT_USERNAME as the default profile. Read-only: does not log in, join groups, scrape members, or send messages.
Expert guidance for building rich text editors with Tiptap - a headless, framework-agnostic editor built on ProseMirror. Use when creating custom nodes, marks, or extensions for Tiptap, implementing input rules or paste rules, working with the Tiptap commands API, building React integrations with useEditor, extending existing extensions, or creating custom node views.
Use when integrating or debugging tiun — the web SDK (@tiun/sdk) or the React Native SDK (@tiun/react-native-sdk) — for authentication, subscription billing, one-time purchases (including repeatable purchases and limited access), webhooks, and time-based paywalls. Triggers on imports of `@tiun/sdk` or `@tiun/react-native-sdk`, calls like `tiun.init`, `tiun.checkout`, `tiun.login`, `tiun.start`, `tiun.listProducts`, `tiun.getProduct`, `tiun.listProductAccess`, `tiun.checkProductAccess`, `TiunProvider`, `useTiun`, `useTiunEvent`, or questions about tiun products, product catalogs, pricing, entitlements, access checks, one-time purchases, lifetime access, referenceId, repeat purchases, limited access, webhooks, fixed fees, time-based sessions, checkout return deep links, or server-side verification.
Set up, run, and contribute to TokHub (github.com/yaojingang/TokHub) — an open-source AI API relay monitoring, recommendation, and OpenAI-compatible gateway system with L1/L2/L3 channel health probing, usage metering, alerts, audit, and Docker self-hosting. Use when the user asks about TokHub, "AI API 中转站监控", cloning/running the Go + React monorepo (TOKHUB_ROLE, sqlc, TimescaleDB, NATS), the L1/L2/L3 probe algorithm, the OpenAI-compatible `/gateway/v1/*` endpoint, or contributing a PR to TokHub. Do not use for connecting a running agent to a live TokHub instance's own API (that is covered by the project's own bundled `agent-skills/tokhub` skill inside the TokHub repo, not this one).
Provides guidance for PyTorch-native agentic RL using torchforge, Meta's library separating infra from algorithms. Use when you want clean RL abstractions, easy algorithm experimentation, or scalable training with Monarch and TorchTitan.
One-off image description using the official `@transloadit/node` CLI. Use `image describe --fields labels` for object-style labels, or `image describe --for wordpress` for structured alt text, title, caption, and description JSON.
Use when writing or changing frontend component code, page layouts, forms, client state, responsive behaviour, animation or styling, or when working in a codebase that uses a design system or component library such as shadcn/ui, Material UI, Ant Design, Chakra UI or Mantine
Reduces and embeds high-dimensional data with umap-learn (UMAP) in Python, including 2D/3D visualization, supervised and semi-supervised UMAP, DensMAP, AlignedUMAP, Parametric UMAP (Keras), transform() on new data, and inverse transforms. Use when visualizing high-dimensional data as a 2D or 3D embedding, preprocessing features for HDBSCAN clustering, using partial labels to guide an embedding, aligning embeddings across time points or batches, or projecting unseen samples into a trained embedding. Tune n_neighbors, min_dist, n_components, and metric. Not for linear PCA or t-SNE-specific workflows.
Develops native Android and iOS plugins for uni-app including module creation, JavaScript-to-native communication, and plugin packaging for distribution. Use when the user needs to build custom native modules, extend uni-app with native capabilities (camera, Bluetooth, sensors), or create publishable native plugins.
Processes and analyzes tabular datasets too large for RAM using Vaex, a Python library for lazy, out-of-core DataFrames over memory-mapped HDF5 and Arrow files, with CSV and Parquet import/export. Covers virtual columns, filtering, groupby aggregations, large-data heatmaps and histograms, and vaex-ml transformers, PCA, and K-means. Use when opening or converting multi-gigabyte CSV/HDF5/Arrow/Parquet files, computing fast statistics on billions of rows, visualizing massive datasets, building ML pipelines that do not fit in memory, or speeding up slow aggregations with lazy evaluation and delay=True. Prefer polars when data fits in RAM, or dask for cluster-distributed work.
Germline and somatic short-variant calling pipeline for Illumina short-read data. Use when the user mentions BWA, BWA-MEM2, GATK, HaplotypeCaller, Mutect2, VCF, GVCF, VQSR, variant calling, germline, somatic, SNV, or indel calling from FASTQ/BAM.
Turn a video file someone already has (an MP4, a Loom download, a long call or screen recording) into Jams (shareable jam.dev recordings with a transcript), one per issue, cut where the speaker moves to a new problem, each titled and filed in a folder. Use when the user says "upload these to jam", "turn this recording into jams", "split this video by bug", "make these jams instead of mp4s", or has a recording too long for one Jam.
Manage Docker-based local development environments with Warden. Use when setting up PHP/Node projects (Magento, Laravel, Symfony, WordPress, Drupal), importing databases, running shell commands, debugging with Xdebug, or managing Docker services. Triggers on "warden", "local environment", "docker-compose wrapper", "database import", "shell access".
Use when a 3D asset must fit a web scene budget: sourcing, glTF pipeline (KTX2, meshopt, Draco), colour spaces, licences. NOT for scene code (web3d-runtime), motion (web3d-animation) or Quest store assets (xr-dev). Opt out: no web3d, без web3d. Triggers - "compress this GLB" / "сожми 3D-модель".
Build, define, and import React code components into Webflow via DevLink. Use when creating Webflow code components, configuring declareComponent definitions, working with prop types, styling within Shadow DOM, bundling with Webpack, or troubleshooting DevLink imports.
WeChat Official Account publishing pipeline and layout template library. Provides ready-to-use templates for policy explainers, long-term care, and technical long-form articles; compiles Markdown into HTML with inline styles that can be pasted directly into the WeChat Official Account editor. Also covers the platform's hard constraints — title and summary length, non-clickable external links in the body, over-wide code blocks, cover image cropping, and no edits after publishing — with check_mp.py for pre-publish checks. Writing and polishing are delegated to dedicated skills rather than duplicated. Use when the user wants to lay out a WeChat Official Account article, needs templates, wants to write for the account, wants to convert an existing draft into a publishable form, needs a pre-publish check, or asks about the platform's limits.
Logs and tracks machine learning experiments with Weights & Biases (W&B, wandb): metrics, hyperparameters, checkpoints, sweeps, artifacts with lineage, model registry, custom charts and shareable reports. Includes integrations for PyTorch, HuggingFace Transformers, PyTorch Lightning and Keras/TensorFlow. Use when logging training runs and comparing them across configurations, running automated hyperparameter sweeps, versioning datasets and models as artifacts, registering model versions, or sharing results with a team workspace. Use when setting up wandb.init for offline or unstable-connection training. Not for general data versioning outside ML runs.
Meta-skill for designing orchestrator+phases structured workflow skills. Creates SKILL.md coordinator with progressive phase loading, TodoWrite patterns, and data flow. Triggers on "design workflow skill", "create workflow skill", "workflow skill designer".
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.