Skills
101,174 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Plans and audits use of ChicagoHAI HypoGeniC/HypoRefine for LLM-assisted hypothesis generation from labeled text datasets. Use for the `hypogenic` package, its task configs, hypothesis banks, or HypoBench datasets—not for manual hypothesis formulation or scientific validation.
Reasoning skill for medical imaging study design and biomarker selection. Use when the user asks to plan an imaging study, choose a preprocessing strategy, select an imaging biomarker, design a radiomics pipeline, handle DICOM de-identification, plan longitudinal imaging analysis, or pick a registration target. Triggers include "imaging study", "preprocessing strategy", "DICOM de-identification", "imaging biomarker", "radiomics", "longitudinal imaging", "registration target", "MNI vs native space", "scanner harmonization", "ComBat", "IBSI", "multi-site imaging", "burned-in PHI", "test-retest reliability", "volumetric biomarker", "diffusion MRI", "perfusion imaging", "fMRI study design", "spectroscopy biomarker".
Use for bug, regression, broken behavior, or production issue reports that need expected behavior, reproduction evidence, root cause analysis, or a fix plan before code changes. Trigger when the report is incomplete, high-impact, user-facing, or likely tied to recent changes. Do not trigger for new feature work, security scan remediation, documentation-only requests, or tiny direct fixes where the user explicitly asks to patch now.
Use for product feature work that explicitly needs discovery, a PRD, gated implementation planning, or stakeholder approval before code changes. Trigger for ambiguous, cross-cutting, client-requested, or high-risk functionality changes. Do not trigger for small direct code edits, routine refactors, bug fixes, security scans, documentation-only work, or questions about existing code.
Use to produce or refresh the architectural and functional documentation of an existing codebase (brownfield) so that humans and the incu-way flows understand the application. Invoked by incu-way-init during brownfield onboarding, or standalone to document a module or refresh stale docs. Especially important when the repo has real code but no living architecture docs, or when the existing docs have drifted from the code.
Prepares and structurally reviews readiness evidence for ISO management-system and laboratory-competence standards - ISO 13485 medical device QMS, ISO 14971 device risk management, ISO/IEC 17025 testing and calibration laboratories, and ISO 15189 medical laboratories. Use when organizing declared scope, controlled documents, risk-management files, scope of accreditation, traceability, CAPA, external-provider controls, or bounded local evidence manifests, and when separating ISO certification from laboratory accreditation, FDA QMSR inspection, CLIA certification, MDSAP, and EU MDR/IVDR evidence boundaries. Not for legal applicability, compliance, certification, or accreditation decisions; contains no clause text.
Interactive issue management with menu-driven CRUD operations. Use when managing issues, viewing issue status, editing issue fields, performing bulk operations, or viewing issue history. Triggers on "manage issue", "list issues", "edit issue", "delete issue", "bulk update", "issue dashboard", "issue history", "completed issues".
Search the JLCPCB electronic components database (~7 million parts) for hardware/electronics projects. Use when the user needs to: (1) Find electronic components (resistors, capacitors, inductors, ICs, connectors, diodes, transistors, MOSFETs, op-amps, microcontrollers, sensors, LEDs, etc.), (2) Look up specific part numbers, LCSC numbers (C-prefix), or manufacturer part numbers, (3) Find alternatives or equivalents for components, (4) Check component availability and stock at JLCPCB/LCSC, (5) Get component specifications (package type, footprint, description), (6) Search for parts for PCB assembly (PCBA) projects, (7) Find SMD or through-hole components, (8) Look up voltage regulators (LDO, linear, switching), audio jacks, connectors, or any other electronic parts. Keywords: JLCPCB, LCSC, electronic components, PCB parts, SMT assembly, BOM, bill of materials, component sourcing.
Workflow for learning CuTe Python DSL by reading, importing, profiling, and extracting reusable patterns from CUTLASS Blackwell example kernels. Use when: (1) studying CUTLASS CuTe DSL reference implementations, (2) importing CUTLASS examples into the project runtime infrastructure, (3) building CuTe DSL knowledge base entries from profiling experiments, (4) understanding CuTe DSL API patterns, TMA pipelining, warpgroup scheduling, or persistent kernel structure.
Build taste through volume of output, deliberate failure, and speed. Taste cannot be developed by watching. It's sharpened by making, failing, and making again. Use when designing practice routines, breaking through creative blocks, or helping someone move from analysis to execution.
Conduct legal research and risk analysis using GoodLegal MCP tools. Use this skill whenever the user asks a legal question, wants to research case law or legislation, needs a legal risk assessment, or asks about French or EU law. Trigger on any mention of jurisprudence, legal research, contract risk, regulatory analysis, legal memo, or references to GoodLegal tools — even if the user just says something like "can you look into whether this clause is enforceable" or "what does the case law say about X".
Use when needing to look up book content inside **my own Amazon (Kindle) book library** (purchases + any personal docs), find a book by title/author, download an ebook I own (where Amazon permits), or reference material from a published book I've purchased. Triggers on book lookups, ebook downloads, "find the book", "get the PDF/EPUB of" **from my library**. Downloads produce PDF/EPUB/MOBI files (and may include Kindle formats depending on the title) - use ebook-extractor skill to convert to text.
Write link text that makes sense out of context for screen reader users and improves usability for everyone. Use when writing or reviewing links, calls to action, navigation labels, or any clickable text. Triggers on: link text, click here, read more, learn more, screen reader links, link purpose, anchor text, call to action, CTA, ambiguous link.
Understand a project through its existing memory when starting or resuming work, or when past decisions matter. Use StrictDoc docs/, Magic Context, claude-mem, and project evidence as complementary, overlapping sources. Identify current authoritative norms separately from history, recover relevant context, and surface meaningful contradictions. Choose the depth and retrieval path that suit the task.
Use when translating or localizing UI strings, marketing copy, documentation, or any source text into French (fr). Enforces vous/tu consistency, le/la gender + number agreement, French typography (non-breaking spaces before : ; ? !), guillemets « », anglicism avoidance per Toubon Law, and prepositional-chain simplification.
Fill web forms by fetching form fields from a URL, deep-searching the user's local knowledge base for relevant info, and generating a markdown document with all answers pre-filled. Use when the user provides a URL to a web form (conference application, speaker submission, event registration, profile form) and wants help filling it out from their existing materials. Also trigger when the user mentions "填网页表", "fill web form", "网页填表", "表单填写", "申请表填写", "conference application", "speaker submission", "讲师申请", "报名表", or provides a URL with "form", "feedback", "apply", "register", "submit" in the path.
Generate "e-magazine × e-ink" style horizontal-swipe web PPTs (single HTML file), with templates for WebGL fluid backgrounds, serif headings + sans-serif body text, section title cards, big-number data slides, and image grids. Use when the user wants to make a share / talk / launch-style web PPT, or mentions "magazine-style PPT", "horizontal swipe deck", "editorial magazine", "e-ink presentation".
Mathematics subject expertise for study notes, problem-solving, and explanations. Covers algebra, calculus, statistics, linear algebra, and discrete math. Provides formulas, proof strategies, and step-by-step solutions. Use when studying math topics, creating math notes, solving math problems, or explaining mathematical concepts. Triggers - math help, algebra, calculus, derivatives, integrals, statistics, linear algebra, proofs, equations.
Bandpass, lowpass, bandstop filter design — hairpin, coupled-line, combline, stub, SIW for frequency selection and harmonic rejection. TRIGGER: user asks to design, create, or analyze any RF filter (bandpass, lowpass, highpass, bandstop, hairpin, coupled-line, combline, stub, SIW). Invoke BEFORE writing code — filter class names differ from what you would guess. SKIP: EM simulation/S-parameter extraction of an existing filter (use matlab-analyze-em), general PCB layout assembly (use matlab-assemble-pcb-layout), material/stackup setup only (use matlab-manage-pcb-material), optimization sweeps (use matlab-optimize-pcb-design).
Prepare structured briefings for meetings with legal relevance and track resulting action items. Use when preparing for contract negotiations, board meetings, compliance reviews, or any meeting where legal context, background research, or action tracking is needed.
Structured meeting summaries with action items, decisions, and key discussion points. Use when taking meeting notes, summarizing discussions, tracking action items, or when the user mentions meeting notes, minutes, action items, or needs structured meeting documentation.
Place specific real furniture or decor products into a room photo using MeltFlex AI. Use when the user wants exact items (a particular sofa, bed, table, lamp) staged into a room while matching their real colors, materials, and proportions. Companion to /meltflex:design.
Establish useful memory for a project that already has development history, or improve an incomplete memory setup. Survey the project and its existing evidence, then initialize or enrich human-readable StrictDoc docs/ with current norms and historical context. Use Magic Context and other available memory alongside it where useful. No previous workflow, legacy folder convention, or pre-existing StrictDoc skeleton is assumed.
MISRA C:2025 expert that operates in two modes: (1) Review — scan existing C code for violations across all ~223 guidelines (22 directives + ~201 rules), report findings with rule IDs, corrected code, and deviation justification templates; (2) Develop — generate new C functions, modules, or data structures that are MISRA-compliant from the first line, applying the essential type model, safe control flow, and banned-construct avoidance from the start. Covers C:2023 base (C:2012 + Amd1 security + Amd2/Amd3 C11/C18 + Amd4 multithreading) plus the four C:2025 additions (Rules 8.18, 8.19, 11.11, 19.3), Rule 15.5 disapplication, and 69 refined guidelines. Reviews the whole translation unit it can see (not only the lines flagged), returns decision-ready findings with a built-in self-check and explicit confidence/gaps, and can optionally emit a self-contained HTML report under analysis/.
Provides comprehensive guidance for MySQL database including SQL syntax, data types, indexes, transactions, stored procedures, and optimization. Use when the user asks about MySQL, needs to write SQL queries, design database schemas, optimize MySQL performance, or work with MySQL features.
Best practices for navigation and routing in Flutter using GoRouter, covering hierarchical route trees, type-safe `@TypedGoRoute` definitions, named navigation, redirects, and deep linking setup. Use when creating, modifying, or reviewing routes, deep links, redirects, or navigation logic that uses package:go_router or package:go_router_builder, including widget tests that mock GoRouter or provide it through `InheritedGoRouter`. Also use for call-site navigation, however small: an `onTap`, button, or tap handler that takes the user from one screen to another, choosing between `context.go`, `goNamed`, `context.push`, `pushNamed`, or `pop`, navigating to a path such as `/products/p1`, passing data to a destination, or returning a result to the caller. A one-line navigation call is still navigation work. Route motion belongs to the animations skill, even inside a `GoRouteData` subclass.
Defines concepts, quality criteria, and boundaries by showing what they are NOT -- using anti-goals, near-miss examples, and failure patterns to create crisp decision criteria where positive definitions alone are ambiguous. Use when clarifying fuzzy boundaries, defining quality criteria, teaching by counterexample, preventing common mistakes, setting design guardrails, disambiguating similar concepts, refining requirements through anti-patterns, or when user mentions near-miss examples, anti-goals, what not to do, negative examples, counterexamples, or boundary clarification.
Opinionated house conventions for NestJS backends — module boundaries and layering, provider wiring and injection scopes, guards/interceptors/pipes/filters, DTOs and validation, domain exceptions, config, and the Nest 11 / Express 5 behaviour that compiles cleanly and fails at runtime. Use whenever the code is NestJS (`@nestjs/*` in package.json, or `@Module`/`@Injectable`/`@Controller`/`@Cron` in the file), including when the question sounds like plain TypeScript, an OOP refactor, a code review, or an ops problem — the Nest-specific answer differs from the generic one. Typical asks: scaffolding a module, adding a controller, provider, or endpoint, wiring an enhancer, a boot-time DI or circular-import failure, a response leaking entity fields despite `@Exclude()`, a cron job firing on every replica, upgrading NestJS, or reviewing a Nest diff. Not for plain Express or Fastify, Angular, frontend, SQL, or CI.
Provides comprehensive guidance for NestJS using the official documentation. Use when the user asks about NestJS architecture, controllers, providers, modules, middleware, guards, pipes, interceptors, dependency injection, GraphQL, WebSockets, microservices, OpenAPI/Swagger, security, or testing.
Fresh install of OpenChoreo onto local k3d, an existing Kubernetes cluster (k3s/GKE/EKS/AKS/DOKS/self-managed), or across multiple clusters. Use when the user says "install OpenChoreo", "set up OpenChoreo on k3d", "bootstrap OpenChoreo on my cluster(s)", "spin up OpenChoreo locally", or "install the OpenChoreo platform".
Aike Zhibo Open Platform "vehicle entry and exit" workflow SOP (end-to-end orchestration of entry/exit). Use when the user wants to get a vehicle fully through entry or exit, simulate the real physical entry/exit flow, run the whole chain of "supplementary record/snapshot entry → correction" or "fee query and settlement → exit snapshot correction → automatic release/gate open", or asks "how does a vehicle enter / exit / entry-exit flow / entry-exit SOP / how to get a vehicle into (out of) the parking lot / simulate a vehicle entering and exiting / why did the vehicle not exit after correction / how to open the gate for release", or when a step gets stuck or needs troubleshooting during entry/exit (e.g. the vehicle did not enter/exit after snapshot or correction, snapshot reports 908, correct succeeds but the vehicle does not leave, or you do not know how to continue to the next step). This skill is a cross-domain orchestration layer that connects multiple commands across parking (supplementary record/correction/inventory) + device (snapshot/gate open) + trade (fee query/payment), and explains the hard constraints across commands (paired exit, snapshot device, channel release mode automatic/manual, payment before release, occasional 908, token validity) and failure handling. Boundary: to only query a single record / on-site vehicle / lock vehicle / call a single command, use the corresponding domain skills openydt-record / openydt-device / openydt-billing directly, do not go through this workflow skill.
Enables Flash Attention for transformer models using PyTorch native scaled_dot_product_attention (PyTorch 2.2+) or the flash-attn library, including multi-query attention, sliding window attention, and FP8 on H100 (FlashAttention-3). Covers profiling speedup, checking accuracy against a baseline, and troubleshooting install and GPU support errors. Use when training or running transformers on long sequences (over 512 tokens), when standard attention runs out of GPU memory, when attention is the inference bottleneck, when switching a PyTorch model to the flash backend, or when tuning attention on H100 GPUs. Not for CPU inference, V100 GPUs, or sequences under 256 tokens; consider xFormers for other attention variants.
Provides comprehensive guidance for Oracle database including SQL, PL/SQL, functions, performance tuning (AWR/ASH), backup (RMAN), Data Guard, and RAC. Use when the user asks about Oracle, needs to write Oracle SQL, work with PL/SQL, or manage Oracle databases.
Use the public `orca` CLI to operate Orca-managed worktrees, folder contexts, terminals, repos, automations, artifacts, worktree comments, and the browser embedded inside the Orca app. Use when the user says "$orca-cli", "use orca cli", "Orca worktree", "child worktree", "cardStatus", "spawn codex/claude in a worktree", "read/wait/send Orca terminal", "terminal send", "full handoff", "handover", "give this to another agent", "another worktree", "Orca browser", "orca artifacts", "share HTML/Markdown", "public artifact link", or "control the browser inside Orca". Prefer this over raw `git worktree`, ad hoc PTYs, Playwright, or Computer Use when the task touches Orca-managed state. Use Computer Use for browser windows, webviews, or desktop UI outside Orca's embedded browser.
Use this skill when writing B2B cold email — "write me a cold email for this client/product", "build an email sequence for this persona", "our reply rate is low, help me fix this email". Value-first cold email built to earn replies, not just opens - the reader gets something useful (an insight, observation, benchmark, or ungated resource) from the email itself, whether or not they ever reply. Applies the R.E.P.L.Y. framework (Relevance, Empathy, Payoff, Low-friction ask, You-focused), email anatomy and deliverability rules, cold-vs-warm tone, and a 4-touch sequence stack. For a new client or campaign, run the intake in references/questionnaire.md before writing.
Load when an agent needs to manage, install, update, remove, or validate OpenViking skills via the `ov skills` CLI. Trigger on explicit user requests about skill management, when the user mentions `ov skills`, `install skill`, `update skill`, `delete skill`, `validate skill`, or when an agent needs to discover what skills are available on the OpenViking server.
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.