Skills
96,770 packaged capabilities and playbooks for agents. Install any of them with one command.
Data & databases
All-in-one FASTQ QC and adapter trimming. Auto-detects Illumina adapters, filters low-quality reads, corrects paired-end overlaps, emits HTML+JSON QC in one pass. 3-10x faster than Trim Galore/Trimmomatic. First step before STAR, BWA-MEM2, or Salmon.
Help researchers identify and catalog relevant datasets for empirical research. Use this skill whenever a user asks to find data, locate datasets, identify data sources, or asks "what data exists for...", "where can I get data on...", "is there a dataset that...", or similar. Also trigger when the user describes a research question and needs to know what data could support it, asks about data availability for a particular topic or geography, wants to compare data sources, or mentions needing panel data, cross-sectional data, or time-series data for a project. Trigger on mentions of "find data", "data search", "dataset", "data source", "microdata", "administrative data", "survey data", "public-use data", "restricted data", or when the user specifies parameters like time windows, geographic levels, or data frequency in the context of research. Also trigger when the user asks for help downloading, scraping, or structuring a dataset that has been identified.
Fetches Google Trends data including daily trending topics, real-time trends, interest by region, related topics, related queries, and autocomplete suggestions. Use when the user asks about trending searches, search popularity, keyword trends, or Google Trends data.
Apply event study methodology to measure abnormal returns and cumulative abnormal returns (CAR) around corporate or market events. Use this skill when the user needs to quantify the market impact of announcements, design event and estimation windows, or when they ask 'did this event affect stock price', 'how do I calculate abnormal returns', or 'what is the market reaction to this announcement'.
Apply panel data analysis with fixed effects, random effects, and dynamic GMM to exploit longitudinal variation and control for unobserved heterogeneity. Use this skill when the user has repeated observations over time for multiple entities, needs to choose between FE and RE via Hausman test, or when they ask 'how do I control for firm-specific effects', 'fixed or random effects', or 'how to handle endogeneity in panels'.
Use for any question about a codebase, its architecture, file relationships, or project content—especially when graphify-out/ exists, where the question should be treated as a graphify query first. Turns any input (code, docs, papers, images, videos) into a persistent knowledge graph with god nodes, community detection, and query/path/explain tools
Compress-Cache-Retrieve for huge, repetitive array tool output (logs, scan results, time series, event streams) before it enters context. Keeps an informative sample — endpoints, anomalies/change-points, head/tail — drops the redundant rest to a local cache, and leaves a retrievable hash. Use when a tool returns a long uniform JSON array you must read but mostly skim, and the full set is one command away if needed. Lossy-but-recoverable.
Architecture patterns for GDPR Article 5(1)(c) data minimization and Article 25(1) data protection by design. Covers field-level encryption, data masking, aggregation, pseudonymization per Article 4(5), and anonymization per Recital 26. Includes ENISA pseudonymization techniques and a data minimization assessment matrix.
Build, review, debug, configure, migrate, teach, or plan Apache Kafka work from TypeScript with current docs. Prefer @platformatic/kafka as the TS SDK (Producer, Consumer, Admin, bootstrapBrokers, consume streams, ProduceAcks, transactions, cooperative-sticky KIP-429, KIP-848). Use for topics, partitions, consumer groups, rebalances, offsets, acks, idempotence, EOS, compaction, retention, SASL/TLS, Schema Registry concepts, KRaft, Kafka 4.x, kafkajs migration, DLQ/outbox patterns, and broker ops that affect app produce/consume.
Build and operate a Busabase-backed mainland China A-share strategy experiment desk with a bundled local Hono App-in-Skill, evidence-backed strategy research, strategy-level L1/L2/L3 manual maturity approvals, one CNY virtual ledger per strategy, and dated strategy backtest reports. Use when the user invokes $kelly-invest-stock or /kelly-invest-stock, wants to define or compare A-share strategies, inspect a strategy and its virtual book, record research evidence, manually approve strategy maturity, or review dated backtests, drawdown, and contribution to the total book. It never connects to a brokerage, places orders, moves money, or presents generated analysis as personalized investment advice.
Run any pending knowledge-base migration by querying the deterministic migration chain (`migrate status`) and executing each pending step's documented procedure in-host, with every write delegated to the step's deterministic CLI primitives. Use when the node reader, `doctor`, or `init` reports an out-of-date `schema_version` / legacy flat layout and asks you to migrate, or when the user asks to migrate the knowledge base.
Use when the user wants an autonomous ML research loop that does more than blindly try changes. After every training run the agent analyses what actually happened inside the model — gradients, activations, embeddings, errors, data — and grounds the next change in that evidence. A `<literature>` on/off dial adds scientific-literature grounding: off behaves as a pure analysis-first loop; on searches papers, grades the evidence, and implements only what prior work supports. One change per run; loops forever until interrupted. Not for one-off training runs or hyperparameter sweeps.
Guide for modernizing legacy Python 2 scientific computing code to Python 3 with modern libraries. This skill should be used when migrating scientific scripts involving data processing, numerical computation, or analysis from Python 2 to Python 3, or when updating deprecated scientific computing patterns to modern equivalents (pandas, numpy, pathlib).
Real-time blockchain event monitoring with webhooks across EVM, Solana, and Bitcoin. Use when user asks about webhooks, wallet/program/contract activity, transfers, stream addresses, Bitcoin xpubs, replaying blocks, or creating and listing historical stream jobs. NOT for querying historical or current blockchain state - use moralis-data-api instead.
Use this skill for authoritative Eastmoney MX financial data lookups when the user asks for exact, time-sensitive market or company data such as real-time quotes, historical prices, capital flows, valuation metrics, financial statements, company basics, shareholders, executives, or entity relationships. Do not use for news/reports/announcements (use mx-search), condition-based stock screening (use mx-xuangu), watchlist actions (use mx-zixuan), or simulated trading operations (use mx-moni).
Plan, configure, and chain repo-native Nemotron customization steps into single-step or multi-step pipelines: curation, translation, SFT/PEFT (AutoModel or Megatron-Bridge), pretraining/CPT, RL alignment (DPO/RLVR/GRPO/RLHF), BYOB/MCQ benchmarks, checkpoint conversion, ModelOpt optimization, env profiles, and evaluation of trained checkpoints or existing/hosted endpoints. Use when a request names a Nemotron step or workflow, or asks to clean, translate, train, fine-tune, align, convert, optimize, evaluate, or compose these into a pipeline. Do NOT use for frontend/dashboard/visualization work, generic ML advice, billing/access, or non-Nemotron coding tasks.
Analyze local Q&A session JSONL files and produce a complete, auditable English Markdown weekly report over consecutive time windows, with all data required for downstream Feishu document and whiteboard rendering. Use for requests such as "analyze the latest week of sessions," "generate a community Q&A weekly report," or "compare with last week." This skill does not ingest other message sources or create/upload online documents.
Reproduce a paper's method in minimal, direct PyTorch when no official code is available, to check whether the method works as claimed. Use whenever the user says "reproduce this paper", "implement this paper's method", "there's no code for this paper", "I want to check if this paper actually works", "turn this algorithm/equation into code", or shares a paper/arXiv link and wants a working implementation of its core mechanism. The output is a small, single-file, readable implementation plus staged verification gates (shape checks → overfit tiny batch → baseline comparison at toy scale) and an explicit assumptions ledger for everything the paper left underspecified. Do NOT use for designing training recipes for your own experiments (pytorch-training-recipe / jax-training-recipe), auditing an existing released artifact (reproducibility-audit), or structuring a research repo (research-codebase).
Run or resume the complete Standard mathematical-modeling workflow from contest files through reproducible evidence, adaptive formal writing, native-equation Word output, and final render QA. Use as the entry point for complete-paper requests.
Use when the user wants to orchestrate defect image generation, run associated setup, or handle outputs on OSMO. The Day 0 path handles cold-start with USD-to-ROI, image-edit augmentation, and AnomalyGen to create initial PCBA datasets. The Day 1 path performs inference and labeling on real images. This skill helps with first-time asset setup, creation of finetuning checkpoints, and configuring deployment. Trigger keywords: defect image generation, dig workflow, dig pipeline, defect image detection workflow, aoi pipeline, aoi anomalygen, usd2roi anomalygen, day 0 pcba, day 1 pcba, day 1 real-photo alignment, day 1 manual roi, metal surface anomaly, glass defect, anomalygen finetune, setup_pcb, setup_metal, setup_glass, setup_pretrained, dig setup, dig datasets, dig pretrained checkpoint, dig image-edit endpoint.
Use when running video data augmentation and auto-labeling workflows on OSMO: flow selection, preflight, submit-time interpolation, monitoring, and output retrieval. Trigger keywords: video data augmentation, data enrichment, auto labeling, VDA demo, OSMO workflow, pseudo labeling.
Portfolio-level pipeline forecast + health rollup: stage-by-stage conversion vs benchmark, risk-adjusted forecast (commit/best/worst scenarios), probability calibration, and a pipeline report card. Pulls deals from HubSpot via the Epiphan AI MCP (execute_sql / hubspot_search_deals); CSV export is a fallback only. Use when: "analyze pipeline health", "pipeline health", "risk-adjusted forecast", "forecast accuracy", "which stage do deals stall in", "pipeline report card", "scenario planning", "deal slippage". For per-deal momentum scoring use deal-momentum-analyzer-skill; for kill/recover triage use dead-deal-recovery-skill.
Use when writing or modifying Python code that calls polars `DataFrame.join` or `LazyFrame.join` and any downstream step depends on the result's row order — attaching a looked-up column back onto order-sensitive data, positional alignment with a numpy array or `pl.Series`, `hstack`, or reproducible output. Skip for joins whose output you immediately re-sort or treat as an unordered set.
Use when the user is planning a two-arm comparison (an A/B test, a simple RCT, a behavioral study, or a two-model/two-config evaluation) and needs to size it and preregister it before collecting data — finding the per-group sample size that hits target statistical power for the smallest effect worth detecting, auditing the design against a validity checklist, and locking it in a preregistration. Only for a single two-arm comparison with one primary outcome. Not for factorial, repeated-measures, clustered/multilevel, time-series, adaptive, or survival designs; not for analyzing data already collected; not for choosing the outcome or manipulation from domain knowledge.
Pro+ subscription required. Tenant-wide Power Automate monitoring using the FlowStudio MCP cached store: failure rates, run-health trends, maker/app inventory, inactive owners, and compliance/health reports. Use only for aggregated tenant views. For one environment, one flow, run control, or root-cause debugging, use power-automate-mcp, power-automate-debug, or the server monitor-flow bundle. Requires FlowStudio for Teams or MCP Pro+.
Browse, slice, and analyze the curated American Reindustrialization directory — with diffs, geo clusters, sector... Trigger phrases: `look up american reindustrialization companies`, `find reindustrialization jobs`, `browse american manufacturing directory`, `sector heatmap of reindustrialization companies`, `use american-reindustrialization`, `run american-reindustrialization`.
Research competitor products, SKUs, pricing, packaging, positioning, feature comparisons, product catalogs, category pages, retailer listings, marketplace data, and market intelligence. Use when the user asks for competitor SKU discovery, product data enrichment, pricing/spec extraction, product comparison, market intelligence, or crawl/map-driven product research.
Core PyGraphistry workflow: auth, DataFrame-to-graph shaping, and first interactive plot. Use when asked to "register graphistry", "get started with pygraphistry", "plot my edges dataframe", "graphistry.register()", "bind src and dst columns", "make a hypergraph", "materialize nodes", or any first-graph / ETL-to-plot task. Also triggers on "first graphistry graph", "graphistry install", "api=3", or questions about graphistry auth credentials. Proactively suggest when the user is setting up graphistry for the first time or can't get a basic plot working from a DataFrame.
Run SQL against Amazon Redshift — submit statements, poll status, page through results, and browse databases/schemas/tables. Use this whenever the user wants to query Redshift (provisioned cluster or Serverless), ask "what tables are in this schema", check a query's status, or mentions `redshift-data`, a Redshift cluster identifier / workgroup name, or a `redshift-data.{region}.amazonaws.com` endpoint. Always start from this skill when interacting with this service — its bundled scripts and recipes are the fastest path.
Designs and implements data retention schedules compliant with GDPR Article 5(1)(e) storage limitation principle. Maps data categories to retention periods with legal basis justification, regulatory minimum holding periods, and automated review triggers for schedule maintenance. Activate for retention policy, storage limitation, data lifecycle, retention period queries.
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.