Skills
97,501 packaged capabilities and playbooks for agents. Install any of them with one command.
Data & databases
Comprehensive guidance for CQRS Architecture — independent CQRS skill covering L1/L2/L3 implementation levels, Event Sourcing, idempotency design, domain event lifecycle, and architecture-specific CQRS integration patterns for Layered/Onion/Hexagonal/Clean/COLA. Use when user asks about CQRS, 读写分离, Event Sourcing, 事件溯源, Command Bus, Query Model, or domain events.
Debug and support PostHog Experiments (A/B tests) for a customer looking at their own results. Use whenever an experiment support ticket is pasted or a customer asks a results question, most commonly "why aren't my exposures even?", "why is one variant getting no traffic?", "why am I missing / seeing too few exposures?", "why does the bias banner show?", or "why don't PostHog's numbers match my SQL?". Pulls the experiment's real data read-only, matches it to a known-cause catalog, and produces a customer-facing explanation, fix, and review of the pertinent numbers. Loads diagnosing-experiment-health as its deep diagnostic library. DO NOT TRIGGER when: creating an experiment (use creating-experiments), only configuring rollout (configuring-experiment-rollout) or metrics (configuring-experiment-analytics), asking lifecycle questions (managing-experiment-lifecycle), or the underlying feature flag is what's misbehaving rather than the results (use debugging-feature-flags).
Guides developers in selecting and implementing multi-region patterns for CockroachDB applications, covering active-passive vs active-active architectures, REGIONAL BY ROW, GLOBAL tables, manual geo-partitioning with lease preferences, and live demo setup with validation queries. Use when designing multi-region database topologies, choosing between REGIONAL BY ROW and manual partitioning, building multi-region demos, or optimizing cross-region latency.
Before collecting data, turn "what do we want to verify" into a statistically sound, reproducible, interpretable research design. Use when the user wants to design an experiment or study plan, choose a design type (completely randomized / randomized block / factorial / fractional factorial / crossover / repeated measures / cluster randomized / response surface), perform randomization and blocking, estimate sample size and statistical power, plan machine learning ablations or benchmarks, or create a statistical analysis plan and preregistration. Covers only "pre-collection design"; post-collection statistical/omics execution is left to data analysis skills.
Explore the dataset before designing any model. Place `data_analysis/data_analysis.py` and do not execute it. Run `materialize.py` once, write `data_analysis.md` and JOURNAL § Data understanding. Never design the model, edit `src/<pkg>/`, or modify raw data files. TRIGGER when the user asks to explore, profile, or understand the data; triage sent them here (`status.data_analysis` missing); a data source changed; they want to refresh a recorded EDA; or leakage or research arises on a recorded EDA. STOP when `status.setup.pending` is non-empty (load `setup-ml-project`), when there is no data (send to triage), the request is not raw-data exploration, or EDA is recorded with no refresh and no methodology concern. Do not invent a package root or a root `JOURNAL.md` while setup is pending. HOW TO USE: G-TABULAR via `add-python-package`, infer or ask the targets, load `plot-ml-figure` if installed, then ask the five-option continuation board, including Close.
Build and run commands to fetch Japanese stock market data with the J-Quants CLI (jquants). Supports fetching stock prices, OHLC, intraday/minute bars, tick data, margin trading balances, short-selling ratios, futures, options, financial statements, dividends, TOPIX and index data. Also supports checking API availability per plan and the available data period (Free/Light/Standard/Premium). Triggers: jquants, J-Quants, 株価, 株価データ, 銘柄, 四本値, 分足, OHLCV, JPX, 東証, TOPIX, 日経225, 先物, オプション, デリバティブ, 空売り, 信用取引, 配当, 財務諸表, 決算, 決算発表予定日, PER, PBR, ROE, EDINET, 大株主, 政策保有株式, 大量保有報告書, 有価証券報告書, 指数, バルクダウンロード, 日本株, Japanese stock, stock price, market data, financial statements, プラン, サブスクリプション, subscription, plan, Free, Light, Standard, Premium, 取得可能期間, データ範囲, data range, API availability, 更新時刻, データ更新, 更新タイミング, データが古い, 最新データ, いつ更新, data freshness, update schedule, update time, when is data updated, latest data
The KubeSense MCP tool layer — connection and auth, the full 38-tool inventory, tool selection, the discovery-first rule, the catalog-label field contract shared by every logs/traces query, WHERE syntax, multi-datasource formula queries, and how to read the TSV/columnar output formats. Read this when a tool returns a field-name or WHERE error.
Generates a comprehensive crypto market report using CoinMarketCap MCP data. Use when users ask about overall market conditions, sentiment, or want a summary. Also use for questions about fear/greed, BTC dominance, altcoin season, trending narratives, or general "how's the market" queries. Trigger: "market report", "market overview", "what's happening in crypto", "market sentiment", "fear and greed", "is it altcoin season", "/market-report"
Design and validate digital filters in MATLAB. Use when cleaning up noisy signals, removing interference, filtering signals, designing FIR/IIR filters (lowpass/highpass/bandpass/bandstop/notch), or comparing filters in Filter Analyzer.
Empirically validates a software metric before trusting or optimizing it — point it at any candidate metric (a command that takes a path and prints one number) plus a corpus, and it runs experiments that try to falsify each property a good metric must have. Checks determinism (same input, same number across runs and hash seeds), invariance to cosmetic edits (also an anti-gaming probe), monotonicity under construct-increasing edits, discrimination, robustness on edge inputs, near-linear tractability, and construct validity (convergent, discriminant vs LOC, predictive AUC, lift over a baseline). Trigger whenever someone proposes, reviews, tunes, or ships a metric, score, or index, asks "is this metric any good", suspects a score tracks LOC or jumps between runs, or builds a deterministic optimization target. It is the empirical companion to the deterministic-metric-design skill and is read-only.
Use when reading or editing .sql files in an mz-deploy project (a directory with project.toml) — resolve object references, columns, and types with the LSP tool instead of grepping. Covers go-to-definition, hover, and document/workspace symbols for Materialize SQL.
Guides initial CockroachDB cluster provisioning and production deployment. Self-Hosted covers cockroach start/init, Kubernetes deployment (Operator, Helm), hardware sizing, and production configuration. Advanced/BYOC covers Cloud Console, API, and Terraform provisioning with production settings. Standard covers cluster creation and provisioned compute selection. Basic covers cluster creation and spending limits. Use when creating a new cluster, preparing for production go-live, or validating deployment configuration.
Core mechanics of the Rerun Chunk Processing API (rerun.chunk) — LazyChunkStream pipelines, Chunk, lenses (MutateLens/DeriveLens/Selector), RrdReader, and writing optimized RRDs. Read BEFORE writing any ingestion/conversion/preprocessing code (convert an MCAP, build a recording from a dataset, preprocess an .rrd, port an old converter): it mandates reader+lens pipelines and steers away from hand-built chunks — no Chunk.from_columns for data a reader/lens can produce, no per-message rr.log, no manual pa.array assembly. Source-specific knowledge lives in the importer skills (rerun-mcap, rerun-urdf, rerun-parquet, rerun-mp4, rerun-lerobot); read rerun-data-model first to decide what the data should become
How raw multimodal robot data maps onto the Rerun data model. Read FIRST, before modeling or converting a dataset - and whenever you are about to convert/ingest/preprocess robot data into an .rrd or build a Rerun recording, even if not asked for the data model. Resolves the entity-vs-component, property-vs-component-vs-layer, and static-vs-temporal decisions and routes to the mechanism (do it with readers and lenses, not hand-built chunks or per-message rr.log): rerun-chunk-processing and the importer skills rerun-mcap, rerun-urdf, rerun-parquet, rerun-mp4, rerun-lerobot.
The industry standard library for machine learning in Python. Provides simple and efficient tools for predictive data analysis, covering classification, regression, clustering, dimensionality reduction, model selection, and preprocessing.
Content extraction from Chinese news sites. Supports WeChat Official Accounts, Toutiao, NetEase News, Sohu News, and Tencent News. Activate when the user needs to extract Chinese news content, scrape WeChat Official Account articles, crawl news, or get news as JSON/Markdown.
Predict construction project costs using Machine Learning. Use Linear Regression, K-Nearest Neighbors, and Random Forest models on historical project data. Train, evaluate, and deploy cost prediction models.
Performs comprehensive due diligence on a cryptocurrency using CoinMarketCap MCP data. Use when users ask about a specific coin beyond just its price. This includes questions like "what is [coin]", "is [coin] legit", "analyze [coin]", tokenomics questions, holder distribution, or any request for deep information about a single cryptocurrency. Trigger: "research [coin]", "tell me about [coin]", "should I invest in [coin]", "DYOR [coin]", "is [coin] safe", "/crypto-research"
Read, create, edit, and delete the cube definitions behind an org's RevOS semantic model, through the RevOS MCP server. Use this whenever the user wants to change what the semantic model contains, even if they never say "cube": model a new table or source stream ("make HubSpot tickets queryable"), add or fix a dimension or measure ("add a revenue sum", "the close date should be a time field"), join two cubes ("link deals to companies"), rename or remove a field, delete a cube, see how a cube is defined ("show me the SQL behind crm_deals"), or find out why the semantic model or a table suddenly fails to load. Read it before any api_write or api_delete on the `cubes` resource. For answering business questions from the data, use semantic-model instead; for audiences, use segments.
Q&A reference for the DataStage parallel engine — parallelism, partitioning theory, APT configuration files, concurrent job execution, restart/recovery, disk/resource tuning, dataset performance, flow optimization (partitioning/sorting/memory), and per-stage semantics. Use for conceptual engine questions and stage property lookups regardless of authoring tool.
Configures and enforces password policies on CockroachDB clusters including minimum length, complexity requirements, and hash cost settings. Use when strengthening authentication requirements, setting up password policies for a new cluster, or meeting compliance password standards.
Guide users to build charts and data visualizations using the Graphic Flutter library. Use when users ask how to create charts, configure chart options, add interactivity, customize appearance, or need help with any aspect of the Graphic charting library.
Use when handling database errors in Frappe/ERPNext. Covers DuplicateEntryError, LinkValidationError, MandatoryError, TimestampMismatchError, CharacterLengthExceededError, InReadOnlyMode, QueryTimeoutError, SQL injection errors, frappe.db.sql parameter format (% vs %s), get_value returning None, transaction deadlocks, MariaDB gone away, too many connections. Error-to-fix mapping for v14/v15/v16. Keywords: database error, DuplicateEntryError, TimestampMismatchError,, MariaDB error, MySQL error, column not found, table missing, duplicate entry, database crash. SQL injection, deadlock, MariaDB gone away, query timeout.
Use when creating or modifying DocType JSON definitions, choosing fieldtypes, configuring naming rules, adding child tables, or setting up tree structures. Prevents invalid DocType configurations from wrong fieldtype choices, broken naming rules, and misconfigured child table links. Covers DocType JSON schema, all fieldtypes and their properties, autoname/naming_rule patterns, child table (Table fieldtype), tree DocTypes, virtual DocTypes, Single DocTypes. Keywords: DocType, fieldtype, naming_rule, autoname, child table, tree, virtual DocType, Single, JSON definition, Custom Field, Customize Form, add field without code, hierarchy, tree view, field types list.
DeFi ecosystem analysis via Gate-Info MCP. Use this skill whenever the user wants TVL rankings, protocol metrics, yield/APY, stablecoins, bridges, exchange reserves, or liquidation heatmaps. Trigger phrases include Uniswap TVL, DeFi ranking, USDC yield, bridge volume, exchange BTC reserves, liquidation density. Route coin-only fundamentals to gate-info-coinanalysis. Sub-scenario tools: info_platformmetrics_* family and info_coin_get_coin_info (exact tool names per scenario in SKILL.md).
History of Market (historyofmarket.com) — public API with 88 historical datasets of US indices since 1871. S&P 500 (price, CAPE, EPS, drawdowns, changes, constituents), Nasdaq Composite/Nasdaq 100 (price, volatility, VXN, changes), Dow Jones, SOX/SMH, sector ETFs (XLK, XLF), Magnificent 7 and macro. No API key, CORS open, CC BY 4.0.
TAM/SAM/SOM market sizing with top-down and bottom-up estimation methods, cross-validation of assumptions, and divergence reconciliation. Generates investor-ready materials with growth projections and confidence intervals. Use when estimating addressable markets, validating opportunity size, or preparing pitch deck market slides.
Terraform modules for deploying self-managed Materialize on AWS (EKS), Azure (AKS), and GCP (GKE). Covers networking, Kubernetes clusters, managed PostgreSQL, object storage, the Materialize operator, and Materialize instance configuration. Use this skill whenever the user asks about deploying self-managed Materialize with Terraform, setting up Materialize infrastructure on any cloud, configuring EKS/AKS/GKE for Materialize, troubleshooting a self-managed deployment, upgrading Materialize instances, customizing Terraform variables for Materialize, or working with the materialize-terraform-self-managed repository. Also trigger when the user mentions Karpenter node pools for Materialize, persist or metadata backend URLs, the Materialize operator Helm chart, CRD versions (v1alpha1 vs v1), rollout strategies, or NVMe swap configuration for Materialize nodes.
Use when modifying EXISTING Datex Studio reports on a branch. Handles label/style changes, field rearrangement, adding/removing columns, datasource modifications, and adding new data sections. Trigger for: "edit a report", "modify a report", "change the label", "add a column", "update the report on branch X", "fix the report layout". For creating NEW reports from scratch, use `report-creator`.
Comprehensive guide for SciPy - the fundamental library for scientific and technical computing in Python. Use for integration, optimization, interpolation, linear algebra, signal processing, statistics, ODEs, Fourier transforms, and advanced scientific algorithms. Built on NumPy and essential for research and engineering.
Entry point for agents and developers new to ScraperAPI. Read first when integrating ScraperAPI into an agent or app, setting up the MCP server for Claude Code or other LLM tools, choosing the right ScraperAPI product, or obtaining an API key. Trigger on: "how do I use ScraperAPI", "set up ScraperAPI", "connect ScraperAPI to Claude", "which ScraperAPI product should I use", "get web data for my agent", "scrape a website in my app", "add web scraping to my project", "ScraperAPI getting started". After reading, route to the narrower skill that owns the selected path. Note: Transmits user-supplied queries, URLs, and content to ScraperAPI.
Use when writing Python that processes biological sequences (DNA/RNA/protein) with the seqpro package — encoding, one-hot, k-mer shuffling, reverse complement, GC content, variable-length sequence batches, or anything involving seqpro's `Ragged` array. Covers the seqpro API surface and the conventions you need to use it correctly.
Create, read, edit, and delete RevOS tables (scoring models), and scope their rows with a segment, through the RevOS MCP server. A table's rows come from a cube, and its columns are read from cubes, stored in RevOS, or filled by actions. Use this whenever the user wants a table or list in RevOS ("make a table of our companies", "a table of the stalled deals"), wants to add, change, rename, hide, or remove a column, pull a field from a related object into a table ("add the company's industry to the deals table"), add an action column, see how a table is built, or delete a table, even if they never say the word "table". Read it before the first api_write or api_delete on the `tables` or `table-views` resource.
Sync columns from a synced data warehouse table onto PostHog person or group properties, so warehouse data becomes usable anywhere person and group properties already work: feature flag targeting, cohorts, insight filters and breakdowns, surveys, session replay filters, workflows, and the person profile. Use when the user wants to "add a person property from my warehouse", "enrich people with Stripe/Postgres/Salesforce data", "put ARR or plan tier on my persons", "target a feature flag by a warehouse column", "sync warehouse columns onto groups or organizations", or wants to inspect, backfill, disable, or debug an existing warehouse-backed person or group property.
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.