Skills
97,501 packaged capabilities and playbooks for agents. Install any of them with one command.
Data & databases
Diagnoses bias, anomalies, and strange results on a PostHog experiment. Covers 0-exposure experiments, sample ratio mismatch, identity fragmentation, multi-variant exposure, uneven-split exclusion bias, significance traps (peeking, A/A, Bayesian vs Frequentist), PostHog-vs-SQL discrepancies, surprises after mid-run edits, and qualitative follow-up via a variant-split survey. TRIGGER when: user asks 'is my experiment biased?' or 'why 0 exposures?', references the bias banner, says a variant looks strange / wrong / off, sees significance flipping or A/A significance, finds PostHog numbers disagreeing with their SQL, reports surprises after mid-run edits, or wants qualitative feedback or a survey for an experiment. DO NOT TRIGGER when: creating an experiment (use creating-experiments), only configuring rollout (use configuring-experiment-rollout) or metrics (use configuring-experiment-analytics), or only asking lifecycle questions (use managing-experiment-lifecycle).
Use this skill when querying, filtering, or summarizing business data in HAP worksheets with the hap command line—especially when filters are complex, require multiple AND/OR conditions or nested groups, or when creating pivot table aggregations (sums/counts/averages/grouping dimensions). Trigger whenever the user says “find records in a table that meet … conditions,” “filter data by status/date,” “how do I write this filter,” “calculate monthly/category totals,” or “create a pivot/summary,” even if they don’t explicitly name the tool. Not for writing data (use the record command to add, delete, or update records).
When the user wants to build data enrichment workflows, score leads against ICP, set up Clay waterfalls, or improve contact data quality. Also use when the user mentions “enrichment,” “data enrichment,” “Clay,” “waterfall enrichment,” “ICP scoring,” “lead scoring,” “intent data,” “contact verification,” “Apollo,” “ZoomInfo,” or “data quality.” This skill covers lead enrichment waterfalls, ICP scoring frameworks, and contact verification systems. Do NOT use for technical implementation, code review, or software architecture.
Professional consulting-style data visualization and infographic creation skill for strategic analysis and executive presentations. Create high-density, boardroom-ready visualizations in McKinsey/BCG style with institutional authority and editorial financial-report aesthetics. Covers 11 visualization types (including cover slides), detailed color palettes with hex codes, aspect ratio standards, prompt engineering templates, and strategic framing guidance for insight-driven deliverables.
General-purpose Scrapy Cloud skill — deploy projects, schedule spiders, list or stop jobs, and view items or logs. Use when asked to deploy a project or spider to Scrapy Cloud / Zyte Cloud / Scrapinghub, schedule or run a spider remotely, manage jobs, or inspect scraped items and logs
Implement the Syncfusion WPF Maps (SfMap) control for geographical data visualization in Windows Presentation Foundation applications. Use this when displaying geographical data, rendering shape files, or integrating map layers with markers and bubbles. This skill covers ShapeFileLayer, ImageryLayer, map customization, legends, OpenStreetMap/Bing Maps integration, and interactive map features.
Implement and configure the Syncfusion WPF SmithChart (SfSmithChart) component for visualizing impedance and admittance in transmission line applications. Use this when working with high-frequency circuit analysis, RF engineering, or telecommunications data. This skill covers series configuration, dual-axis systems, data markers, legends, appearance customization, and interactive features.
Guide for implementing Syncfusion WPF Surface Chart (SfSurfaceChart) for three-dimensional data visualization. Use this when working with 3D surface charts, contour charts, wireframe charts, or mathematical surface plotting with X, Y, Z coordinates. This skill covers surface visualization, topographical charts, and 3D data plotting scenarios for WPF applications.
Use this skill when the user asks to add documentation, add docs, add references, or install documentation about Neon. Adds Neon best-practice reference links to project AI documentation (CLAUDE.md, AGENTS.md, or Cursor rules). Does not install packages or modify code.
Use when implementing the data layer in Android or KMP — the Repository pattern's layered error-propagation model (repository as the error boundary, sealed DataError, Result placement with and without a domain layer) and Room in commonMain (@ConstructedBy, BundledSQLiteDriver, per-target KSP).
Load this skill whenever the project contains charts, graphs, data visualizations, infographics, or any visual representation of data (bar charts, line charts, pie charts, scatter plots, dashboards). Under no circumstances render a chart without a text alternative, data table, or summary. Prioritize conveying the same information through non-visual means.
API reference for CoinMarketCap cryptocurrency endpoints including quotes, listings, OHLCV, trending, and categories. Use this skill whenever the user mentions CMC API, asks how to get crypto data programmatically, wants to build price integrations, or needs REST endpoint documentation. This is the go-to reference for any CMC cryptocurrency API question. Trigger: "CMC API", "coinmarketcap api", "crypto price API", "get bitcoin price via API", "/cmc-api-crypto"
Copy a PostHog endpoint (a saved HogQL/insight query exposed as an API route) to another project in the same organization, or duplicate it under a new name in the same project. Use when the user wants to duplicate an endpoint, promote an endpoint from staging to production, replicate an endpoint's query/variables/freshness config in another workspace, or clone an endpoint to iterate on it. Unlike feature flags and experiments, endpoints have NO native cross-project copy tool — this skill covers the read-then-recreate flow (endpoint-get then endpoint-create), the active-project switching it requires, name-collision checks, and the safe defaults (land unmaterialised in the target, verify with endpoint-run). Does not cover editing endpoint versions (see managing-endpoint-versions) or authoring a brand-new endpoint from scratch (see creating-an-endpoint).
Expert customer health scoring and analytics guidance. Use when designing health scores, building churn prediction models, analyzing usage metrics, identifying at-risk accounts, creating executive dashboards, or performing cohort analysis. Use for leading indicator development, customer data enrichment, risk escalation frameworks, and retention analytics.
Salesforce Data Cloud integration patterns and architecture (2025). PROACTIVELY activate for: (1) Data Cloud setup and ingestion, (2) Data Streams (cloud, mobile, web SDK, ingestion API), (3) data model objects (DMO) and source objects (DSO), (4) identity resolution and unified profiles, (5) calculated insights and segmentation, (6) activations to Marketing Cloud, advertising platforms, Salesforce CRM, (7) Bring Your Own Lake (BYOL) with Snowflake, BigQuery, Databricks, (8) zero-copy data sharing, (9) Data Cloud + Agentforce grounding, (10) consent management and compliance. Provides: data-stream selection matrix, identity resolution rules, segmentation patterns, BYOL configuration, and activation playbook.
Decision frameworks for DatoCMS content modeling — schema shape, field choice, content reuse, taxonomies, content vs presentation, admin UI organization. Use for modeling *decisions*, not implementation: model vs block; single_block vs Modular Content vs Structured Text; references vs embedded blocks; taxonomy shape (flat/tree/faceted); refactoring page-shaped schemas to reusable content; fitting 300 KB / 500-block / 5-level record limits; model behaviour (singleton, draft mode, all_locales_required, sortable/tree/ordering_field, presentation_title_field, collection_appearance, inverse_relationships_enabled); field config (validator + appearance — enum + string_select, slug auto-fill, required_alt_title, structured_text allowlists, framed vs frameless single_block). Also schema review (reuse, editor ergonomics, omnichannel). *Creating* schema → `datocms-cli` or `datocms-cma`. Query/render → `datocms-cda` + `datocms-frontend-integrations`
Evaluate one learner with `skore.evaluate`. The splitter and any non-default score are already on the DataOp from `build-ml-pipeline`. This skill writes the call without `splitter=`, persists the report, and records the locator. When the locked split is the training table and test table that shipped with the data, fit on the training table, then pass `splitter="prefit"` with only the test table. When `model-ml-pipeline` dispatched this turn, return the locator. A standalone turn also owns the evaluate-stage close. TRIGGER when code calls `cross_val_score`, `cross_validate`, `classification_report`, `.skb.cross_validate`, or a handwritten metric print, or the user asks to score, evaluate, or run CV on one learner. Narrative reads of a persisted report belong to `audit-ml-pipeline`. HOW TO USE: before any evaluation call. Resolve G-SKORE-MODE, read the stops, and emit Pre-flight before code. Confirm symbols with `python -m skore_skills api get`.
Generates code that fine-tunes a base model using SageMaker serverless training jobs. Use when the user says “start training”, “fine-tune my model”, “I'm ready to train”, or when the plan reaches the fine-tuning step. Supports SFT, DPO, RLVR, and RLAIF trainers, including RLVR Lambda reward function and RLAIF custom prompt creation.
Guide for using the Grafana MCP to monitor and diagnose the Node.js ingestion pipeline workers in production. Use when investigating event lag, drops, pipeline errors, person/group processing, Kafka consumer health, Redis, Postgres, ClickHouse downstream health, or any ingestion worker question. Covers prod-us and prod-eu environments.
Use when setting pythia up for a project or a machine - writing connections.json, creating a least-privilege agent user, wiring the SQLcl MCP server, or diagnosing a connection that will not open. Also the place to start when check prints a privilege warning.
Comprehensive guide for implementing Syncfusion WPF Sparkline (SfSparkline) controls in Windows Presentation Foundation applications. Use this when working with sparklines, mini charts, or trend visualization. This skill covers sparkline types (line, column, area, WinLoss), markers, track ball, range bands, axis controls, and segment customization for compact data visualization in WPF applications.
The Pelosi Tracker. Watches the stock trades members of Congress must disclose under the STOCK Act: reads the House Clerk's public Periodic Transaction Reports (PTRs) as soon as they are posted, alerts once per new filing, and turns each PDF into clean rows (who, owner, stock, ticker, buy or sell, options strike and expiry, trade date, amount range, how many days it took to disclose). Flags trades in stocks you hold. Senate reports are covered by a documented manual route. Use when the user says "what is Pelosi buying", "track Nancy Pelosi", "Congress trades", "politician stock trades", "did any congressman buy X", "alert me when Pelosi files", or "run the Pelosi skill".
Guide for adding a new RPC to personhog-replica and personhog-router. Covers eligibility checks, proto definition, code generation for Python and Node.js clients, Rust implementation (storage trait, postgres queries, service handler, router wiring), and index compatibility validation. Use when adding a new gRPC endpoint to personhog, migrating a Django ORM query to personhog, or extending the personhog service API.
Build automated IOC enrichment pipelines in Splunk Enterprise Security by ingesting threat feeds into KV Store collections and correlating them against security events via lookup tables, modular inputs, and the Threat Intelligence Framework. Use when wiring threat intel into Splunk correlation searches to flag IOC matches and reduce SOC triage time.
Simplified database tool for metadata queries, SQL execution, nl2sql and similar scenarios on Volcano Engine RDS instances and self-built databases. Use this skill when the user needs to query table structures of Volcano Engine RDS instances and self-built databases, view data, execute SQL, or convert natural language to SQL.
Generates code that transforms datasets between ML schemas for model training or evaluation. Use when the user says “transform,” “convert,” “reformat,” “change the format,” or when a dataset's schema needs to change to match the target format—always use this skill for format changes rather than writing inline transformation code. Supports OpenAI chat, SageMaker SFT/DPO/RLVR/RLAIF, HuggingFace preference, Bedrock Nova, VERL, and custom JSONL formats from local files or S3.
Query Feishu attendance data (user-task / user-stats). user-task checks clock-in tasks/shifts; user-stats checks attendance statistics (attendance/late arrivals/early departures/leave). Note: Only Tenant Token is supported; SDK v3.5.3 does not accept User Token. Each query is limited to 31 days; longer ranges are rejected. Use when the user asks to “check attendance,” “check clock-in records,” “attendance statistics,” or “attendance details.”
Analyzes and correlates live Infrahub data via the MCP server — answers operational questions, detects drift, and investigates impact. TRIGGER when: querying infrastructure data, checking compliance, investigating change impact, producing ad-hoc reports. DO NOT TRIGGER when: writing automated checks, building transforms, designing schemas, populating data files, planning an Infrahub version upgrade (use infrahub-planning-upgrades). ALWAYS pass the user's question verbatim as args — this skill runs in a forked context and cannot see the parent conversation. Invoking without args will fail.
Diagnose and improve one existing functional Splunk search from supplied SPL and runtime evidence. Use when a search, report, dashboard panel, or scheduled search is slow, queued, expensive, resource-intensive, or prematurely finalized and the user needs evidence-backed query tuning, acceleration-fit analysis, workload separation, or a comparable before-and-after plan. Route new-search authoring, functional break/fix, governance, and deployment-wide operations to their owning workflows.
Fit a Datadog LLM-Obs evaluator to human labels. Takes an annotation queue, works out where in the trace the labelled property actually lives, drafts an LLM-judge that predicts the human label, scores that judge against the already-labelled rows with a metric agreed with the user, then hill-climbs it — inspect the errors, make one focused change, re-score, keep it only if it beats the best — for a bounded number of iterations, and finally publishes the winner to Datadog as a DISABLED evaluator (not a Datadog draft — a real evaluator with `enabled: false`). Use when the user says "build an eval from my annotations", "build an evaluator from the annotation queue", "turn my annotations into an evaluator", "learn an evaluator from my labels", "fit a judge to the annotation queue", "auto-label", "auto labelling", "automate this annotation queue", "scale up my human labels", or wants the rest of a queue graded the way the humans graded the first rows. Needs an annotation queue with at least two classes present in the human labels (e.g. one true and one false for a boolean).
Manage the full lifecycle of Alibaba Cloud E-MapReduce (EMR) ECS clusters—creation, scaling, renewal, and status queries. Use this Skill when users want to set up big data clusters, view cluster status, add nodes, release nodes, configure auto-scaling, check cluster and node states, or diagnose creation failures. Also applicable for scenarios like "create a Hadoop cluster", "data lake cluster", "running out of resources", "check my cluster", "renew", etc. NOTE: This Skill does NOT support cluster deletion, release, or termination under any circumstances. Any request to delete or terminate a cluster will be refused and redirected to the EMR console.
Google Tag Manager expertise. Audit GTM containers, design tag and trigger architecture, debug firing issues, implement consent mode v2, and advise on server-side tagging. Use when the user asks about GTM, tag management, container audits, consent mode, server-side tagging, or tracking implementation.
Evaluate Omni AI accuracy using Omni's built-in eval system — define a prompt set, run a judged eval against a model (or branch), and read the accuracy-judge verdicts. Use this skill whenever someone wants to evaluate Omni AI, benchmark Blobby, run regression tests, compare AI output across branches or model-context changes, measure AI quality, run A/B tests on model changes, assess the impact of an ai_context or modeling change, or any variant of “run evals”, “test Blobby”, “benchmark query generation”, “compare AI results”, “regression test”, “how accurate is the AI”, or “measure the impact of my changes”.
Protein-protein interaction (PPI) network analysis — STRING (predicted + experimental), BioGRID (curated), SASBDB (small-angle scattering), Complex Portal (curated complex membership and stoichiometry, includes CORUM data). Distinguishes physical interactions (binding) from functional associations (co-expression, co-regulation). Use for interactome queries, complex partner identification, curated complex composition lookup, and pathway-level interaction analysis
Manage the full lifecycle of ADBPG Supabase projects. Use for listing or querying projects, creating, pausing/resuming, resetting passwords, managing API Keys, and security IPs. Triggers: “Supabase,” “supabase project,” “spb-xxx,” “ADBPG Supabase”
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.