Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Data & databases
GPU-accelerates scientific Python on NVIDIA hardware and verifies that the result is correct and faster. Use for CUDA/GPU optimization; CPU-bound NumPy, SciPy, pandas, scikit-learn, NetworkX, scikit-image, vector-search, image-processing, graph, simulation, or file-I/O workloads; CuPy, cuDF, cuML, cuGraph, cuVS, cuCIM, KvikIO, Warp, Newton, Numba-CUDA, or RAFT questions; and profiling, memory-transfer, kernel, or multi-GPU bottlenecks. Also use when large data-parallel Python code is slow and GPU acceleration is a plausible option, even if the user does not name CUDA.
Analyze parquet files using Python and Ibis. Use when the user wants to explore, transform, or analyze parquet data files, perform aggregations, joins, or export results. Works with local parquet files and provides database-agnostic data operations.
Computational pathology toolkit for analyzing whole-slide images (WSI) and multiparametric imaging data. Use this skill when working with histopathology slides, H&E stained images, multiplex immunofluorescence (CODEX, Vectra), spatial proteomics, nucleus detection/segmentation, tissue graph construction, or training ML models on pathology data. Supports 160+ slide formats including Aperio SVS, NDPI, DICOM, OME-TIFF for digital pathology workflows.
Fetch and analyze protein structures from RCSB PDB. Use this skill when: (1) Need to download a structure by PDB ID, (2) Search for similar structures, (3) Prepare target for binder design, (4) Extract specific chains or domains, (5) Get structure metadata. For sequence lookup, use uniprot. For binder design workflow, use binder-design.
Run the PAIDF Orchestration Image Attribute Augmentation DAG on Kubernetes - person-crop clothing augmentation, attribute search, and augmented dataset generation. Select for requests about image attribute augmentation, person attribute search, person re-identification data, clothing augmentation, attribute captions, augmentation payloads, run status, or result retrieval. Runs environment setup first when controller readiness is unknown. Not for video or defect-image generation.
Design Power BI report layouts, select chart types, plan page structures, choose themes, and apply data storytelling BEFORE generating PBIR JSON. Produces a Design Spec plus the official canonical `Design Brief:` contract (design identity, page archetypes, mechanical layout_contract) that feeds `power-bi-report-authoring`. Covers the `design` mode of Microsoft's `powerbi-report-cli` (tone, signature, archetype router, chart selection, color, typography, layout, interactivity, accessibility, brownfield redesign, pre-flight checklist). Use whenever the user asks to design a report, plan a dashboard layout, choose visualizations or chart types, select a theme or palette, plan navigation, redesign or critique an existing report, or needs a design spec from requirements. Do NOT use to generate PBIR JSON, validate report structure, or build the semantic model (use `power-bi-report-authoring` / `power-bi-semantic-model`).
Use this skill for reinforcement learning tasks, including high-performance RL training, custom environment development, vectorized parallel simulation, multi-agent systems, or integration with existing RL environments (Gymnasium, PettingZoo, Atari, Procgen, etc.). Use it to implement PPO training, create PufferEnv environments, optimize RL performance, or develop policies with CNNs/LSTMs.
Python library for working with DICOM (Digital Imaging and Communications in Medicine) files. Use this skill when reading, writing, or modifying medical imaging data in DICOM format, extracting pixel data from medical images (CT, MRI, X-ray, ultrasound), anonymizing DICOM files, working with DICOM metadata and tags, converting DICOM images to other formats, handling compressed DICOM data, or processing medical imaging datasets. Applies to tasks involving medical image analysis, PACS systems, radiology workflows, and healthcare imaging applications.
Solves single- and multi-objective optimization problems in Python with pymoo, using NSGA-II, NSGA-III, MOEA/D, SPEA2, RVEA, GA, DE and PSO. Covers custom problems (Problem, ElementwiseProblem, FunctionalProblem), constraint handling, mixed-variable problems, ZDT/DTLZ/WFG benchmarks, genetic operators, parallel evaluation, Pareto front visualization and multi-criteria decision making. Use when finding Pareto-optimal trade-offs between conflicting objectives. Use when defining a constrained or mixed-variable problem and choosing an evolutionary algorithm. Use when benchmarking algorithms on standard test problems. Use when customizing crossover or mutation operators. Use when picking one solution from a Pareto front. Not for gradient-based or convex solvers.
Python interface to OpenMS for mass spectrometry data analysis. Use for LC-MS/MS proteomics and metabolomics workflows including file handling (mzML, mzXML, mzTab, FASTA, pepXML, protXML, mzIdentML), signal processing, feature detection, peptide identification, and quantitative analysis. Apply when working with mass spectrometry data, analyzing proteomics experiments, or processing metabolomics datasets
[Deprecated] This is the required documentation for agents operating on the CloudBase Relational Database through MCP. It defines the canonical SQL management flow with `queryMysqlDatabase`, `manageMysqlDatabase`, `queryPermissions`, and `managePermissions`, including MySQL provisioning, destroy flow, async status checks, safe query execution, schema initialization, and permission updates. New environments should use PostgreSQL — see postgresql-development skill instead.
Runs a full forensic audit of outbound REPLIES (positive responses, objections, ghosts) for an agency client and writes the result as a structured page in an established 9-section format. Use whenever someone shares a batch of outreach replies and wants them analyzed — "analyze these replies", "audit the replies", "what's wrong with our responses", "why are leads ghosting", "analyze positive responses" — or links a Google Sheet / CSV / Notion table / mentions a LinkedIn-automation or cold-email campaign of replies. Accepts ANY source (Google Sheet, CSV, Notion table, raw pasted text, or an outreach-platform MCP). Always diagnoses ROOT CAUSE (targeting/data problem vs SDR/pitch problem), classifies every reply, extracts gold-standard reply templates from the data, and ships a 9-section audit. NOT for benchmarking a single metric ("is my reply rate good") — that's outbound-analyst.
Covers classical machine learning in Python with scikit-learn (sklearn): classification and regression estimators, clustering and dimensionality reduction, preprocessing, Pipeline and ColumnTransformer, cross-validation, metrics, and GridSearchCV hyperparameter tuning. Use when training or comparing classifiers and regressors on tabular or text data, clustering data and choosing the cluster count, building leakage-free preprocessing pipelines, evaluating models with cross-validation and metrics, or tuning hyperparameters. Not for deep learning; use a dedicated neural network framework instead.
Comprehensive toolkit for survival analysis and time-to-event modeling in Python using scikit-survival. Use when working with censored survival data, performing time-to-event analysis, fitting Cox models, Random Survival Forests, Gradient Boosting models, or Survival SVMs, evaluating survival predictions with concordance index or Brier score, handling competing risks, or implementing any survival analysis workflow with the scikit-survival library.
This skill should be used when working with single-cell omics data analysis using scvi-tools, including scRNA-seq, scATAC-seq, CITE-seq, spatial transcriptomics, and other single-cell modalities. Use this skill for probabilistic modeling, batch correction, dimensionality reduction, differential expression, cell type annotation, multimodal integration, and spatial analysis tasks.
Statistical visualization with pandas integration. Use for quick exploration of distributions, relationships, and categorical comparisons with attractive defaults. Best for box plots, violin plots, pair plots, heatmaps. Built on matplotlib. For interactive plots use plotly; for publication styling use scientific-visualization.
Language-independent tokenizer treating text as raw Unicode. Supports BPE and Unigram algorithms. Fast (50k sentences/sec), lightweight (6MB memory), deterministic vocabulary. Used by T5, ALBERT, XLNet, mBART. Train on raw text without pre-tokenization. Use when you need multilingual support, CJK languages, or reproducible tokenization.
Email campaign/sequence performance review composite. Pulls campaign data (sends, opens, replies, bounces), reads actual email copy and subject lines, analyzes reply content (objections, positive interest, questions), and produces a diagnostic report covering quantitative metrics, copy quality, lead quality, and actionable recommendations. Tool-agnostic — works with Smartlead (MCP), Instantly, Outreach, Lemlist, Apollo, or CSV data.
Initialize and configure Astro/Airflow projects. Use when the user wants to create a new project, set up dependencies, configure connections or variables, or understand project structure. For running the local environment, see managing-astro-local-env
Parse a SIF (Standard Interchange Format) input into a readable preview and optionally append canonical product rows to the project's CSV library. Use when asked to convert or inspect SIF dealer data. For the reverse direction use /as:csv-to-sif.
Runs a signal-research pass over an EXISTING firmographic account base (CSV, Google Sheet, Notion DB, CRM export) for a B2B service agency: decides which signals to hunt for this ICP, detects them account-by-account from cheapest source upward, records evidence URL + date for every hit, scores and heat-tiers the accounts, flags expired signals for re-scan, and returns an enriched list plus a coverage report that says which sources actually produced signals. Use when the user says: "знайди сигнали по базі", "прогони базу на сигнали", "ресерч сигналів", "signal research", "у нас є список компаній — хто зараз гарячий", "перескануй базу, сигнали вигоріли", "which accounts have a live trigger right now", "enrich this list with signals", or hands over a company list and asks who to contact first. Requires a base — this skill does NOT build the list (data-research, niche-data-finder) and does NOT invent the signal library (that lives in `resources/signals-catalog.md`). It also does not write outreach copy.
Use when writing SpacetimeDB Unreal Engine clients, generated Unreal bindings, Unreal plugin setup, `DbConnection`, ticking the connection, cache access, subscriptions, reducer invocation, or identity handling. Triggers on: Unreal, Unreal Engine, UE, Unreal SDK, URemoteTable, USubscriptionBuilder.
Provides comprehensive technical analysis for stocks and ETFs using RSI, MACD, Bollinger Bands, and other indicators. Activates when user requests stock analysis, technical indicators, trading signals, or market data for specific ticker symbols.
Wrap JSON-backed database columns with ActiveModel-like classes using store_model. Use when creating configuration objects, managing nested JSON attributes, or adding validations to JSON data. Triggers on JSON columns, store_model, typed JSON, configuration objects, or nested attributes.
Run Supabase Management API SQL for persistent data tasks such as querying records, applying schema changes, managing policies, and handling storage metadata. Use when requests involve Supabase database CRUD, migrations, or production-like data inspection.
Designs and implements Tilebox scene-catalog ingestors from STAC, provider XML or JSON, object-storage prefixes, and COG file trees. Use when onboarding a new metadata source, designing its canonical schema and collections, normalizing records to STAC 1.1 semantics, authoring Python or Go converters, constructing Assets, Links, Storage, or Authentication messages, or validating and ingesting converted datapoints.
Analyze and visualize time spent on software projects by combining data from multiple sources: WakaTime coding time, git commit session detection, Claude Code usage, Codex CLI usage, and Cursor IDE usage. Produces both an interactive HTML dashboard (dark-themed, Chart.js) and a Markdown report with ASCII charts. Use when the user asks to: analyze work hours, calculate time spent on a project, generate a work hours report, visualize coding activity, create a project time breakdown, or summarize development effort across date ranges.
Graph-based drug discovery toolkit. Molecular property prediction (ADMET), protein modeling, knowledge graph reasoning, molecular generation, retrosynthesis, GNNs (GIN, GAT, SchNet), 40+ datasets, for PyTorch-based ML on molecules, proteins, and biomedical graphs.
Audits the creator's own Instagram account through Treg - installs Treg if needed, connects their Instagram, pulls the profile, the last 30 to 50 reels with their real stats (views, reach, average watch time, saves, shares, comments), and the account's reach over 28 days, finds what over-performs against their own median, looks at the first seconds of the best and worst reels, and writes video-agent/audit.md with what works, what doesn't and the 3 changes to make. Use when the user says audit, audit my Instagram, analyse my account, my stats, why my reels don't work, what works on my account, my best reels, Instagram insights, or connect my Instagram.
Use when the user is building on Web 3 Data Spaces (W3DS) or the MetaState prototype — building a post-platform, integrating an eVault, calling the eVault GraphQL API (createMetaEnvelope, updateMetaEnvelope, removeMetaEnvelope, bulkCreateMetaEnvelopes, uploadFile, bindingDocument*), wiring the Web3 Adapter, writing a webhook controller for /api/webhook, authoring mapping.json files, using the wallet-sdk, implementing the w3ds://auth or w3ds://sign flow, resolving W3IDs / eNames via the Registry, working with the Ontology service, proposing a new ontology, dealing with Binding Documents, dereferencing w3ds://file URIs, provisioning an eVault, syncing public keys, hosting a platform on GitW3 (the W3DS-aware Git forge — `.w3ds/platform.json`, platform eName, version eName, PPA certification, deployment records, porting an existing app, `w3ds-deployment-key.json`), or debugging local dev (Registry, Provisioner, eVault-core, Dev Sandbox, pnpm dev:core). Also use for any design decision about where data lives on a W3DS platform — adding a table, entity, model or field, deciding what to cache locally, or asking whether something belongs in the database or the eVault. Also use for company signing delegation — a company eVault with directors, titled roles, delegations and re-delegations, signing on behalf of a company, verifyDelegatedSignature / buildDelegatedSignRequest / buildGrantSignRequest. Also use when the user asks what an eVault, W3ID, eName, MetaEnvelope, Envelope, Ontology, Web3 Adapter, Awareness Protocol, or Awareness-as-a-Service is.
Turns company domains into named decision-makers with verified work emails, or finds emails for contacts you already have — a people-search node keyed on domain, then a multi-source email cascade with a known cost per contact. Use when asked to enrich a list, find contacts or emails, work from LinkedIn URLs, or raise find rate without guessing patterns. Triggers include "заенрич", "знайди персон", "потрібні пошти", "find contacts", "who do we write to at these companies".
Builds a canonical WEEKLY outreach report for a client as a standalone narrative page — the exact 7(+2)-section format instead of a dashboard dump. Pulls live data from the client's outreach platform (LinkedIn automation and/or email sending tools), applies the known platform gotchas, classifies every reply with root cause, and ends with gold leads + ready-to-send drafts. Trigger with: "weekly report", "outreach report", "analyse this week's results", "build the report from our template". NOT for batch reply forensics alone (use reply-audit) and NOT for benchmarking a single metric (use outbound-analyst).
Use when writing or reviewing CQL code, schema definitions, or application code that targets YugabyteDB's Cassandra-compatible YCQL API (port 9042). Triggers on YCQL tables, partition keys, TTL, batching, or any mention of YugabyteDB with Cassandra/CQL.
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.