Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Data & databases
Search, scrape, and interact with the web via the Firecrawl CLI. Use this skill whenever the user wants to search the web, find articles, research a topic, look something up online, scrape a webpage, grab content from a URL, get data from a website, crawl documentation, download a site, or interact with pages that need clicks or logins. Also use when they say "fetch this page", "pull the content from", "get the page at https://", or reference external websites. This provides real-time web search with full page content and interact capabilities — beyond what Claude can do natively with built-in tools. Do NOT trigger for local file operations, git commands, deployments, or code editing tasks.
This skill should be used at the start of any computationally intensive scientific task to detect and report available system resources (CPU cores, GPUs, memory, disk space). It creates a JSON file with resource information and strategic recommendations that inform computational approach decisions such as whether to use parallel processing (joblib, multiprocessing), out-of-core computing (Dask, Zarr), GPU acceleration (PyTorch, JAX), or memory-efficient strategies. Use this skill before running analyses, training models, processing large datasets, or any task where resource constraints matter.
Run the Anti-Autoresearch integrity-forensics sweep (span-anchored evidence ledger → GPT auditors propose findings → a rules-only reporter that lists every proposal with what the auditor said about it) against a paper via a SHA-pinned thin launcher — then convert the verdict into a typed policy gate (BLOCK/WARN/NO_NEW_BLOCKER) and an append-only obligations ledger. Use when user says \"integrity forensics\", \"forensic audit this paper\", \"投稿前自查诚信\", \"审这篇论文的诚信\", or says \"anti-autoresearch\" when the upstream repo's own skills are not installed. Also invoked by /paper-writing (submission self-forensics, default ON), /peer-review (forensic appendix), /resubmit-pipeline.
Analyze a Materialize environment via the MCP Developer endpoint, and/or configure an MCP client (Claude Code, Cursor, VS Code, Zed, Continue, Windsurf, Claude Desktop) to connect to the materialize-developer server. For analysis: check environment health, investigate performance, troubleshoot stale materialized views, diagnose memory pressure, audit resource utilization, run EXPLAIN ANALYZE on user objects, get optimization recommendations. For client connection: configure/connect/set-up an MCP client to materialize-developer (Emulator, Cloud, or self-managed), control which user/role is used, switch between identities. Trigger even if user just says "check my environment", "why is my MV stale", "why is my cluster slow", "what can I optimize", "explain analyze my materialized view", "how do I connect Claude Code to materialize-developer", or "configure Cursor for the Materialize MCP".
Guides building on Qdrant Edge, the embedded in-process shard. Use when someone asks “how to sync Edge with the server”, “keep a local shard in sync with Qdrant Cloud”, “BM25 or keyword search on Edge”, “hybrid search on Edge”, “embeddings on device”, “Edge snapshots”, “apply a partial snapshot”, “why is my Edge search empty after inserts”, or is writing custom sync, BM25, or fusion code against qdrant-edge. Also use when deciding what Edge ships built-in versus what you must implement.
Designs and implements production-grade RAG systems by chunking documents, generating embeddings, configuring vector stores, building hybrid search pipelines, applying reranking, and evaluating retrieval quality. Use when building RAG systems, vector databases, or knowledge-grounded AI applications requiring semantic search, document retrieval, context augmentation, similarity search, or embedding-based indexing.
Find unused and rarely queried metrics in VictoriaMetrics using the metric_names_stats API, then suggest optimization actions (drop rules, relabel configs). Use this skill when the user wants to find unused metrics, identify wasted storage, optimize metric ingestion, reduce cardinality by dropping unneeded metrics, clean up scrape targets, or asks about which metrics are never queried. Also trigger when the user mentions "metric cleanup", "unused series", "what metrics can I drop", "metric optimization", "wasted ingestion", or wants to reduce VictoriaMetrics resource consumption by eliminating unnecessary metrics.
Use when reviewing or rebalancing direct vs. partner-led channel economics—calculating fully loaded cost to serve per channel, channel ROI through cash, LTV, and marginal lenses, and the optimal channel mix subject to constraints. For Heads of Commercial, RevOps, and VPs of Sales conducting quarterly channel reviews when the pipeline is mixed (e.g., 60% direct + 40% partner-led) and nobody knows which channel makes money after accounting for CAC, support load, partner discounts, deal velocity differences, retention differentials, and overhead allocation. Outputs cost to serve, channel ROI verdicts (DOUBLE-DOWN / MAINTAIN / DEFUND / EXIT), a sensitivity-tested channel-mix recommendation, and the diminishing-returns inflection (e.g., “which channel actually makes money—direct or partner?”)
Build RAG / unstructured-document evaluation datasets and demo documents (e.g. for Knowledge Assistant) on Databricks: generate synthetic PDFs locally, upload to Unity Catalog volumes, and pair each document with test questions for retrieval evaluation.
Push records from apps, services, devices, or scripts directly into Unity Catalog Delta tables with Zerobus Ingest: SDKs (Python, TypeScript, Go, Java, Rust, C++, .NET), REST, OTLP, MQTT, Arrow Flight, or Apache Kafka-compatible producer APIs, with no message bus to run. Use to stream events, telemetry, logs, IoT data, CDC, or columnar batches into Databricks in near real time; connect existing collectors and log shippers (OpenTelemetry, Debezium, Vector, Telegraf, Cribl); replace a message bus that only feeds the lakehouse; or debug Zerobus auth, schema, throughput, or delivery issues.
Record-level CRUD and bulk operations — create, update, delete, upsert, CSV import, multi-table foreign-key loads, AI-generated sample data. Use when the user wants to write, modify, seed, or import data records into Dataverse tables, or plan or perform Finance and Operations (ERP) record writes and DMF package imports.
Migrate existing third-party SDK tracking code (Firebase, Amplitude, Sensors Data, Mixpanel, GA4) to the AE SDK (数数 / ThinkingAI / Agentic Engine / AE SDK) — switch (replace) or add alongside existing tracking (dual-write). Trigger words: 埋点迁移、埋点切换、第三方埋点迁移、第三方埋点切换、迁移埋点到数数、把 xx SDK 切换到数数、把已有埋点用数数实现一遍、用数数替换 xx SDK、集成数数并参考已有 xx SDK 埋点、Firebase 迁移、Amplitude 迁移、Mixpanel 迁移、神策迁移、Sensors Data 迁移、GA4 迁移、gtag 迁移、双写、tracking migration、migrate Firebase/Amplitude/Mixpanel/Sensors Data/GA4 to AE、switch xx SDK to ThinkingAI、replace xx SDK with AE、replace logEvent with track、re-implement existing tracking with ThinkingAI、add AE tracking alongside existing analytics、dual-write. Provider-agnostic core with per-provider adapters; reuses ae-generate-tracking-plan and ae-generate-tracking-code.
Set up Critical User Journey (CUJ) monitoring in bitdrift. Deploys a complete observability stack: path discovery (sankey), conversion funnel, completion rate SLO, key step duration alerting, session capture, and a two-tab dashboard. Trigger when a user wants to monitor a journey end-to-end, track conversion and drop-off, measure step duration, or set up SLOs on a business-critical flow such as checkout, onboarding, login, or search.
Best practices for query optimization in MariaDB — indexing strategies, EXPLAIN analysis, pagination, histogram statistics, and MariaDB-specific optimizer settings. Use when diagnosing slow queries, designing indexes, reviewing schema or query performance, or when queries involve large tables, pagination, GROUP BY, or ORDER BY. Also use when the user asks about MariaDB query performance, EXPLAIN output, or optimizer behavior.
Provides guidance for performing causal interventions on PyTorch models using pyvene's declarative intervention framework. Use when conducting causal tracing, activation patching, interchange intervention training, or testing causal hypotheses about model behavior.
Query Tonghuashun Wencai data using natural language in Chinese, including market data for A-shares, indices, funds, Hong Kong and US stocks, and convertible bonds. Use when the user needs to retrieve stock screening, financial, capital flow, technical indicator, or other data from Wencai using natural language.
Translate Cypher and Neo4j-style queries into direct HelixDB v3 Rust SDK requests. Use when the input contains Cypher, Neo4j, MATCH, OPTIONAL MATCH, WHERE, RETURN, ORDER BY, LIMIT, DISTINCT, MERGE, CASE, UNWIND, FOREACH, DETACH DELETE, IS NULL, or variable-length path patterns. When the target is Helix Cloud, always use helix-mcp first.
End-to-end user research assistant — qualitative and quantitative. Use this skill whenever the user mentions user research, user interviews, discussion guides, interview guides, research plans, qualitative research, quantitative research, user surveys, survey design, usability studies, participant recruitment, research synthesis, interview transcripts, research reports, running studies with AI, or explicitly mentions Cookiy AI. Also trigger when users want to talk to customers, conduct discovery research, create a study or survey, analyze interview data, run AI-moderated interviews, or collect survey responses. Covers the full lifecycle: planning studies, creating discussion guides, running AI-moderated interviews (real or synthetic) via Cookiy, designing and distributing surveys, and synthesizing results into reports.
Use when defining database schemas, creating data collections, or setting up structured data storage. Triggers include CMS, collection, database, schema, data model, fields, relationships, permissions, data structure, entity definition, data extension.
Alibaba Cloud SLS (Simple Log Service) index configuration manager skill. Use this skill to help users inspect, create, update, or delete a Logstore index, generate an index configuration from user-provided structured log samples, and optimize an existing index configuration for given query/SQL workloads, write throughput, and storage cost — all through the aliyun CLI. Triggers: "SLS 索引", "索引配置", "create index", "update index", "delete index", "generate index", "optimize index", "全文索引", "字段索引", "log index config", "aliyun sls index".
Designs and implements production-grade ML pipeline infrastructure: configures experiment tracking with MLflow or Weights & Biases, creates Kubeflow or Airflow DAGs for training orchestration, builds feature store schemas with Feast, deploys model registries, and automates retraining and validation workflows. Use when building ML pipelines, orchestrating training workflows, automating model lifecycle, implementing feature stores, managing experiment tracking systems, setting up DVC for data versioning, tuning hyperparameters, or configuring MLOps tooling like Kubeflow, Airflow, MLflow, or Prefect.
Manage Alibaba Cloud DataAnalysisGBI via OpenAPI/SDK. Use whenever the user needs DataAnalysisGBI resource lifecycle operations, configuration changes, status inspection, or troubleshooting for analytics service workflows.
Highload blocks — HighloadBlockTable, compileEntity data classes, HLBLOCK_{id} user fields, hlblock links, iblock directory property, ORM events, hl_element_* rights. Use when creating HL dictionaries or reading/writing their records.
A toolkit of financial time-series analysis methods. Covers stocks, commodity futures, cryptocurrencies, ETFs, forex, and indices, with a complete workflow from data retrieval to advanced analysis. Includes 70+ methods across eight areas: time-series testing, predictive modeling, cross-asset relationships, volatility risk, portfolio optimization, regime identification, commodity-specific analysis, and network analysis. For data retrieval, it prioritizes the tushare MCP tool (A-shares, Hong Kong and US stocks, futures, funds, and macro data); assets not covered by tushare, such as commodity futures (CL=F) and crypto (BTC-USD), are supplemented with yfinance scripts.
Guide for using the Pinecone CLI (pc) to manage Pinecone resources from the terminal. The CLI supports all index types (standard, integrated, sparse) and all vector operations—unlike the MCP, which only supports integrated indexes. Use for batch operations, vector management, backups, namespaces, CI/CD automation, and full control over Pinecone resources
Query the public REST APIs that power dashboard.internetcomputer.org. Covers canister metadata, ICRC ledger data, SNS data, ICP ledger, and network metrics with cursor-based pagination. Use when fetching canister info, token data, SNS proposals, or network stats via HTTP from off-chain code. No canister deployment or cycles needed
Datos macro y sociales de Argentina via la API oficial Series de Tiempo del Estado (apis.datos.gob.ar/series). ~4250 series del INDEC + BCRA + Min Economia + Sec Trabajo. Sin auth, sin API key. IPC nacional, EMAE, IPI, ISAC, EPH (desempleo), pobreza, comercio exterior, salarios (RIPTE, SMVM), tipo de cambio, reservas, REM expectativas. Transformaciones builtin (% YoY, % YTD, change) y agregacion temporal (daily→monthly→yearly) server-side. La API mas estable y mejor documentada del repo.
Performs pandas DataFrame operations for data analysis, manipulation, and transformation. Use when working with pandas DataFrames, data cleaning, aggregation, merging, or time series analysis. Invoke for data manipulation tasks such as joining DataFrames on multiple keys, pivoting tables, resampling time series, handling NaN values with interpolation or forward-fill, groupby aggregations, type conversion, or performance optimization of large datasets.
Turns a plain-English description of a recurring report into a real, repeatable report — defines the metrics, pulls them from connected data sources, delivers a chat summary plus an XLSX workbook, and saves the definition so it can rerun on demand or on a schedule. Handles everything from “sales by location versus last year” to AR aging and labor as a percentage of revenue. Works fully from an uploaded CSV or XLSX when no connector is available. Use whenever the owner asks for a report, dashboard, KPI pack, metrics summary, or recurring numbers — including “can you track this every week,” “I need to see these numbers monthly,” “build me a report on,” “same report as last time,” or “put together a KPI dashboard.” Use it even when the owner names the metrics without saying “report.”
Primary Python tool for 40+ bioinformatics services. Preferred for multi-database workflows: UniProt, KEGG, ChEMBL, PubChem, Reactome, QuickGO. Unified API for queries, ID mapping, pathway analysis. For direct REST control, use individual database skills (uniprot-database, kegg-database).
A Douyin similar-account recommendation tool. Enter a Douyin account name or account ID to retrieve the account’s data, content data, and similar-account recommendations through the Honghu API, then analyze commonalities, differences, and optimization suggestions. Use when the user mentions “recommend similar Douyin accounts,” “find Douyin benchmarks,” “similar Douyin accounts,” “Douyin competitor analysis,” or “find Douyin peers.”
Half-Quadratic Quantization for LLMs without calibration data. Use when quantizing models to 4/3/2-bit precision without needing calibration datasets, for fast quantization workflows, or when deploying with vLLM or HuggingFace Transformers
Query integrated indexes using text with Pinecone MCP. IMPORTANT: This skill ONLY works with integrated indexes (indexes with built-in Pinecone embedding models like multilingual-e5-large). For standard indexes or advanced vector operations, use the pinecone-cli skill instead. Requires the PINECONE_API_KEY environment variable and a configured Pinecone MCP server
This skill should be used when working with annotated data matrices in Python, particularly for single-cell genomics analysis, managing experimental measurements with metadata, or handling large-scale biological datasets. Use when tasks involve AnnData objects, h5ad files, single-cell RNA-seq data, or integration with scanpy/scverse tools.
Comprehensive toolkit for developing with the CocoIndex library. Use when users need to create data transformation pipelines (flows), write custom functions, or operate flows via CLI or API. Covers ETL workflows for AI data processing, including embedding documents into vector databases, building knowledge graphs, creating search indexes, and processing data streams with incremental updates
Environment-level Dataverse administration — bulk delete, retention/archival, organization settings, OrgDB settings, recycle bin, audit, and the 37 allowlisted PPAC toggles. Use when the user wants to clean up data at scale, configure audit, change environment settings, manage retention policies, or list/cancel ERP batch jobs.
Translate Gremlin and TinkerPop-style traversals into direct HelixDB v3 Rust SDK requests. Use when the input contains Gremlin, TinkerPop, g.V, g.E, hasLabel, has, out, in, both, outE, inE, repeat, emit, dedup, valueMap, count, range, or limit. When the target is Helix Cloud, always use helix-mcp first.
Configures Neon Serverless Driver for Next.js, Vercel Edge Functions, AWS Lambda, and other serverless environments. Installs @neondatabase/serverless, sets up environment variables, and creates working API route examples with TypeScript types. Use when users need to connect their application to Neon, fetch or query data from a Neon database, integrate Neon with Next.js or serverless frameworks, or set up database access in edge/serverless environments where traditional PostgreSQL clients don't work.
Fetch and extract content from web pages, converting HTML to clean markdown. Use when users want to read web articles, extract information from URLs, scrape web content, or when the built-in WebFetch tool fails due to network restrictions. Trigger when user provides URLs to read, asks to fetch web content, or needs to extract text from websites.
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.