Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Data & databases
Build backends and APIs with Directus headless CMS. Use when a user asks to create a headless CMS, build a content API without coding, set up a backend admin panel, create REST or GraphQL APIs from a database, manage content with roles and permissions, build a data platform with auto-generated APIs, or replace traditional CMS with a headless solution. Covers data modeling, auto-generated REST/GraphQL APIs, roles/permissions, flows (automation), file storage, and SDK integration.
Raw reference for ESPN's undocumented public JSON endpoints, for a sport, league, or operation that no dedicated sports-skills skill covers, or when a developer explicitly asks to work with raw ESPN endpoints. Prompt-only: endpoint catalog, documented league slugs, observed response shapes, pitfalls, and stdlib fetch helpers. Output is provider-native ESPN JSON, not normalized or canonical data. Use when: the requested league or operation is genuinely not covered by a dedicated skill (e.g. MMA, lacrosse, rugby, Australian football, NCAA volleyball, field hockey, water polo, IndyCar/NASCAR), or the user explicitly asks to write, debug, or inspect code against ESPN endpoints. Don't use when: a dedicated skill covers the league and job — nba-data, wnba-data, nfl-data, nhl-data, mlb-data, cfb-data (NCAA football), cbb-data (NCAA men's basketball), football-data (its listed soccer leagues), tennis-data (ATP/WTA), golf-data (PGA/LPGA/DP World), cricket-data, fastf1 (Formula 1); sports-news or sports-reporter for news and articles; betting, markets, kalshi, or polymarket for odds and prediction markets. volleyball-data covers Dutch Nevobo volleyball only, not NCAA. A dedicated skill's error, timeout, or empty result is not permission to switch to raw ESPN automatically — report it and ask first
Create, inspect, and edit Microsoft Excel workbooks and XLSX files with reliable formulas, dates, types, formatting, recalculation, and template preservation. Use when (1) the task is about Excel, `.xlsx`, `.xlsm`, `.xls`, `.csv`, or `.tsv`; (2) formulas, formatting, workbook structure, or compatibility matter; (3) the file must stay reliable after edits.
Facebook's library for efficient similarity search and clustering of dense vectors. Supports billions of vectors, GPU acceleration, and various index types (Flat, IVF, HNSW). Use for fast k-NN search, large-scale vector retrieval, or when you need pure similarity search without metadata. Best for high-performance applications.
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
Any live-web task via the Firecrawl CLI — including ordinary web research: searching the web, reading or extracting pages, gathering sources, discovering site URLs, bulk extraction, downloading a site, change alerts, or pages needing clicks/login — web only; local files route to firecrawl-parse. For papers use firecrawl-research-index; for library, API, error, or bug questions use firecrawl-developer-index.
This skill helps users automatically extract structured news data from Google News via BrowserAct API. Agent should proactively apply this skill when users express needs like searching for news about a specific topic, tracking industry trends, monitoring public relations or sentiment, collecting competitor updates, getting latest reports on specific keywords, monitoring brand exposure in media, researching market hot topics, summarizing daily industry news, tracking media activities of specific individuals, retrieving hot events from the past 24 hours, extracting structured data for market research, monitoring global breaking news.
Apply Grounded Theory (Glaser and Strauss) to build theory inductively from qualitative data through open, axial, and selective coding. Use this skill when the user needs to develop new theory from data rather than test existing hypotheses, conduct theoretical sampling and constant comparison, determine when theoretical saturation is reached, or when they ask 'what theory explains this phenomenon', 'how do I code qualitative data systematically', or 'when do I stop collecting data'.
Manage CloudRobo model training tasks and simulation reinforcement learning (SimRL) tasks — create pretrain (TRAIN_FROM_SCRATCH) and finetune (MODEL_TUNING) tasks with FFT/SFT/LORA/QLORA/DEEPSPEED methods; manage the full task lifecycle (create/read/update/delete/stop/restart/resume/draft); save and resubmit draft configs; count tasks by status; monitor execution stages, resource usage, training logs, signed URLs, and events; diagnose failures. SimRL tasks (simulation reinforcement learning) are managed via the same CLI with a --sim-rl flag and cover the same CRUD + lifecycle + monitoring surface (except resume, which is train-only). Triggers include: model training, fine-tuning, pretraining, training task, training stages, resource usage, training logs, training events, draft task, restart training, stop training, resume training, task stats, simulation reinforcement learning, SimRL, 仿真强化学习, 模型训练, 模型微调, 训练任务, 训练阶段, 资源使用, 训练日志, 训练事件, 草稿任务, 重启训练, 克隆训练, 停止训练, 续训训练, 任务统计.
Jimmer ORM development assistant that helps write entity mapping code and DSL queries. Use for: (1) defining or modifying Jimmer entities (interfaces annotated with @Entity); (2) writing Jimmer DSL query code; (3) configuring entity associations (@ManyToOne, @OneToMany, @ManyToMany); (4) using dynamic predicates, dynamic table joins, or implicit subqueries; (5) editing Java/Kotlin files containing Jimmer-related code.
Web crawling and URL discovery with katana. Use when you need to crawl all pages and endpoints of a target site, discover hidden API paths, endpoints in JavaScript, or extract forms. katana is a next-generation crawling framework from ProjectDiscovery that supports standard mode and headless browser mode and can parse JavaScript to discover dynamic endpoints. This skill should be used for any scenario involving web crawling, URL discovery, JS endpoint extraction, form discovery, or sitemaps
Analyzes an app's codebase and cross-references Google Search Console, PostHog, and Google Keyword Planner to identify low-competition keyword footholds and track expansion into adjacent terms. This skill should be used when the user says "keyword wedge", "find keyword opportunities", "seo analysis", "keyword strategy", "find search wedges", "keyword research for my app", "grow organic traffic", "what keywords should I target", "SEO for my app", "organic search strategy", or "how to rank higher". Generates markdown and HTML reports and maintains state across runs for expansion tracking.
Instrument a launch so the traffic spike is ATTRIBUTABLE, not a vanity number — one consistent UTM scheme on every launch link, the conversion funnel (visit → signup → activation) wired into GA4 or the project's existing analytics, and a validate + read-out step. Writes the UTM map + event plan to `.ulpi/launch/<channel>/analytics.md`, parameterized by launch source (e.g. producthunt, hackernews) so each link and report is correctly attributed. Shared measurement block of the launch-* family — invoked by the per-platform runners (`launch-product-hunt`, `launch-hacker-news`, `launch-x`, `launch-linkedin`) to close the loop, or run standalone. Measures durable outcomes (signups, activation), extends the real stack, and NEVER puts PII in links or events. Use when a launch needs UTMs on every link and signup tracking, not just an upvote count.
Analyze AMS waveform data using Mixed-Signal Blockset utilities: phase noise measurement, clock jitter, anti-aliased resampling, timing measurements, lock time, INL/DNL, ADC/DAC calibration, HSpice import. Use when analyzing time-domain voltage from PLL/VCO/clock simulations, measuring phase noise from variable-step solver output, computing jitter, or resampling non-uniform data.
Connect to and acquire images from cameras in MATLAB using Image Acquisition Toolbox. Use this skill when the user wants to: use a webcam, acquire images, capture video, stream from a camera, connect to a GigE Vision camera, preview live video, list available cameras, configure camera properties, use a CoaXPress frame grabber, or use a Camera Link frame grabber. Covers USB webcams, GigE Vision, USB3 Vision, CoaXPress, Camera Link, and GenICam-compliant cameras. Always prefer videoinput from Image Acquisition Toolbox over webcam support package or gigecam.
Build a metrics framework for any product, team, or business. Use when asked for a metrics tree, KPI framework, North Star metric, AARRR funnel, HEART framework, or OKR metrics. Produces a structured metrics hierarchy from North Star down to leading indicators, with measurement guidance.
This skill should be used when the user asks to adapt pretrained, foundation, language, vision, multimodal, or embedding models. PROACTIVELY activate for: (1) transfer learning, full fine-tuning, frozen backbones, adapters, LoRA, QLoRA, AdaLoRA, PEFT, (2) Hugging Face Transformers, Diffusers, Accelerate, TRL, RLHF, DPO, preference tuning, alignment, (3) dataset preparation, instruction tuning, chat templates, tokenization, packing, catastrophic forgetting, (4) RAG, vector databases, embedding optimization, reranking, (5) multimodal and edge fine-tuning. Provides: safe, efficient fine-tuning and adaptation guidance.
Manage refresh schedules, selected-lineage execution definitions, and job execution for existing Microsoft Fabric Materialized Lake Views (MLVs) via REST APIs. Configure, update, and delete schedules and execution definitions; trigger on-demand refreshes; monitor job status; and cancel running jobs. Use `spark-authoring-cli` to author MLV definitions. Discovery, visual lineage, and data quality metrics require the Lakehouse UI. Triggers: "schedule MLV refresh", "manage MLV refresh schedule", "trigger MLV refresh", "MLV job status", "cancel MLV refresh", "monitor MLV refresh", "recurring MLV refresh", "daily MLV refresh".
Query and analyze scholarly literature using the OpenAlex database. This skill should be used when searching for academic papers, analyzing research trends, finding works by authors or institutions, tracking citations, discovering open access publications, or conducting bibliometric analysis across 240M+ scholarly works. Use for literature searches, research output analysis, citation analysis, and academic database queries.
A chemical toolbox designed to speak the many languages of chemical data. Supports over 110 formats and provides tools for conversion, 3D structure generation, molecular searching (SMARTS), and force field calculations. Use for chemical file format conversion (SDF, PDB, SMILES, CIF, Gaussian), 3D coordinate generation from 2D structures, substructure searching with SMARTS patterns, molecular docking preparation, force field minimizations (UFF, GAFF, MMFF94), molecular fingerprints and Tanimoto coefficients, and batch processing of chemical databases.
Use available Polygres MCP tools or the Polygres CLI to authenticate, operate standard or synchronized projects, import data, apply migrations, configure retrieval, manage Runtime keys and durable operations, and check readiness. Use the CLI for explicit command-line work and as the documented fallback for operations outside the MCP catalog.
Translate a tracking plan into an SDK-specific instrumentation guide. Shows how to make identify, group, and track calls using the target analytics SDK with real template code, architecture guidance, and constraint documentation. Outputs .telemetry/instrument.md. Covers 25 analytics destinations across product analytics, CDPs, product adoption, web analytics, error monitoring, feature flags, and session tools. Use when the user has a tracking plan and needs to know how to implement it with a specific SDK like Segment, Amplitude, Mixpanel, PostHog, Accoil, Google Analytics, Sentry, LaunchDarkly, or via generic HTTP POST. Also use when user asks 'create instrumentation guide,' 'how to implement tracking,' 'SDK guide,' or 'generate implementation guide.'
Build a structured product model by scanning the codebase and talking to the user. Produces .telemetry/product.md — a description of what the product does, who uses it, how value flows, and what entities exist. Use when starting telemetry work on a new codebase, when the user asks to 'model this product,' 'understand this product,' 'what does this product do,' 'map the product,' 'product model,' or when no .telemetry/product.md exists yet. This is the entry point for the telemetry lifecycle.
Python interface to PROJ (cartographic projections and coordinate transformations library). Handles transformations between different Coordinate Reference Systems (CRS) and performs geodetic calculations (distance, area on ellipsoids). Use for coordinate transformations, CRS conversions, geodetic calculations, UTM projections, GPS coordinate conversions, ellipsoidal distance calculations, and spatial reference system operations.
Deep learning framework (PyTorch Lightning). Organize PyTorch code into LightningModules, configure Trainers for multi-GPU/TPU, implement data pipelines, callbacks, logging (W&B, TensorBoard), distributed training (DDP, FSDP, DeepSpeed), for scalable neural network training.
Write, complete, or extend Qlik Sense load scripts (.qvs) for data preparation. Primary use case is preparing flat datasets for ML experiments with Qlik Predict — engineering features, handling train/test splits, and outputting QVD files. Also covers general Qlik load script authoring: completing scripts from starting code, comments, or prompts. Use this skill whenever the user mentions Qlik load script, .qvs files, Qlik data load, QVD output, data preparation for Qlik Predict, or asks for help writing or fixing Qlik script syntax. Also trigger when the user provides a Qlik load script and asks for modifications, extensions, feature engineering, or data transformations. Even if the user just pastes Qlik script code and asks a question about it, use this skill.
World-class systematic trading research - backtesting, alpha generation, factor models, statistical arbitrage. Transform hypotheses into edges. Use when "backtest, alpha, factor model, statistical arbitrage, quant research, systematic trading, mean reversion, momentum strategy, regime detection, walk forward, " mentioned.
CRM data quality, deduplication, enrichment automation, record matching, and data decay management. Use when CRM is full of duplicates, records are outdated or decaying, fields are missing or inconsistent, job titles and industries don't match across records, you're not sure how bad your data quality is, or enrichment keeps overwriting good data. Do NOT use for one-time enrichment of a prospect list (use /sales-enrich), building new prospect lists (use /sales-prospect-list), or ZoomInfo-specific config (use /sales-zoominfo). For platform-specific help, use /sales-zoominfo.
Evaluate research rigor and scientific claims. Assess methodology, experimental design, statistical validity, biases, confounding, and evidence quality using GRADE, Cochrane ROB, and preregistration standards. Use for critical analysis of research papers, evaluating AI/ML study claims, reviewing preprints, or fact-checking scientific assertions in any domain.
Defensive exploratory data analysis for scientific data (CSV, FASTA, etc.). Context-first, human-guided; one plot at a time, ask why before executing, append-only journal per session, scripts with PEP723 and uv run, WebP plots. Use when opening data files for EDA or when the user wants guided scientific data exploration.
Use when the user wants TikTok research or workflow guidance for lead generation, influencer discovery, brand monitoring, competitor analysis, content analytics, trend research, or audience analysis, including account analysis, creator discovery, video inspection, comment scraping, transcript extraction, hashtag or song research, and TikTok Shop or product research.
World-class data engineering skill for building scalable data pipelines, ETL/ELT systems, and data infrastructure. Expertise in Python, SQL, Spark, Airflow, dbt, Kafka, and modern data stack. Includes data modeling, pipeline orchestration, data quality, and DataOps. Use when designing data architectures, building data pipelines, optimizing data workflows, or implementing data governance.
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.