Skills
99,448 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Write or rewrite text in plain English that readers outside the field can understand in one read: use short sentences, active voice, one word per meaning, put conditions before commands, define terms at first use, and avoid AI slop. Use for docs, reports, release notes, error messages, and replies; workflow-axa uses it for every HTML report and delivery message.
Use when the user wants to publish, host, or share AI-generated content at a stable URL on slugjar — JSON feeds, RSS, Markdown notes, YAML configs, plain text, HTML pages, or binary files. Triggers: "post to slugjar", "publish to slugjar", "create a slot", "host this at a URL", "make a feed", "share this as a URL", "slugjar".
Load to carry story work with an author from what they want to a page they accept: reading their intent, setting direction, producing and critiquing the prose, owning the verdict and the critique round limit, recording what settled, and presenting the result.
Shared tooling for the game-design skills: one entry script (studio.py) and one Python package that validate concept records, items, permanent upgrades, enemies, combat calculations, in-run growth builds, stat budgets, enhancement and crafting, currency flow, level encounters, and playtest records, draft items and enemies from budget curves, simulate roguelike meta-progression, check numeric provenance tags in design docs, serve a read-only browser viewer that shows all of a project's design data with validation results, and install a read-only judge agent and data-check hooks for Claude Code. Use when viewing, checking, or simulating game design data, when setting up Claude Code hooks for a game project, or when another skill in this repository tells you to run a check.
Turns a gameplay video (generated by a video model such as MiniMax H3 Max or Seedance, or any clip) into a real, playable WebGL game in Three.js whose every image, 3D model, sound and music track is generated with fal, then proves it: side-by-side frames of the video and the game, and a scenario that plays the game with the keyboard and films its best seconds. Ships scripts to study the video (frames, palette, cuts), run and log fal models, record a playtest at the real frame rate, compare game and video, and find the liveliest window of a take. Use when the user wants a video, an AI gameplay clip or a game concept "turned into a game", "made playable", "recreated in the browser" or "rebuilt in Three.js/WebGL". Not for game engines (Unity, Unreal, Godot) or for copying existing commercial games.
Modernize an existing website (association, shop, practice, old Jimdo, Wix or WordPress site) for someone without prior knowledge. Start the inventory immediately without asking: list all pages, every feature (forms, registrations, appointments, PDFs, prices, galleries, maps, languages) and all content, with photos of the old site in docs/bestand.md. Then ask only for information known to the person, and proceed through planning and design to the new site without losing any feature without their consent. Use the skill for “modernize …”, “modernize our website”, “our site is outdated”, “renew website”, “relaunch”, “move away from Jimdo, Wix or WordPress”, “redo the old site”, and immediately when someone sends only a website address (https://…) with no other words.
Clarify a workflow task's intended behavior before any specification: trace real journeys, callers, failure and compatibility paths in current code, ask only about unresolved user decisions, and record the gate-one behavior agreement. Use for a new substantive goal, a material scope change, or a contradictory or partial answer.
Reconcile a finished or stopped workflow task against its evidence: map every agreed requirement, check, changed file, and review finding to current proof, mark stale or missing proof, then write the delivery record and report. Use when implementation ends or someone asks whether the task is really finished.
Build interactive games using Augmented Reality (AR) and camera-based motion and gesture tracking for Smart TVs, Interactive Flat Panels (IFP), digital whiteboards, or laptop webcams. Students can interact at the front of the classroom without touching a physical screen.
Connect course material modules to the academic book-writing ecosystem (Book-Author-Agent-Skills). Guide lecturers and teachers in compiling teaching materials into drafts of textbooks, reference books, or monographs with official ISBNs.
Cut a Semantic Versioning release in any repository: find the version source, decide MAJOR, MINOR or PATCH from the commits since the last tag, update the manifest, the changelog and any current-release line, then commit and create an annotated tag, leaving the push to the user. Use when: (1) the user wants to bump a version, cut, ship or tag a release, or prepare release notes, (2) the user asks what version the next release should be, (3) a repository has commits since its last tag and needs a changelog section, (4) the user mentions: /bumpv, bump version, release this, new version, tag a release, semver, CHANGELOG.md, Keep a Changelog, chore(release), git describe, annotated tag, --follow-tags, package.json version, Cargo.toml, pyproject.toml, "what version should this be". Carries a version-source table for Node, Rust, Python, Go, PHP, Ruby, .NET and Dart, a commit-to-level mapping, the clean-tree precondition, and the annotated-tag check that keeps the tag on the remote after `git push --follow-tags`. Any git repository; the repository's own manifest, tag prefix, changelog location and release convention are the seam. Not a publisher: it never pushes, never creates a GitHub Release and never runs a package registry publish, and it versions one package, not a monorepo workspace
Use when running coinagerie or npx @coinagerie/cli, printing a codebase's vocabulary, checking or writing a glossary, choosing between --check, --write, and a bare run, reading the glossary block or JSON it printed, reading the missing and stale words named by a failed check, finding which words are still undefined, or wiring a glossary check into CI, a commit hook, or an Nx target.
Use when writing or changing a coinagerie.config.ts, choosing which files are scanned, excluding paths or words, reading another glossary into exclude.words, merging a word family with fold, declaring markdown or JSON glossary destinations, sharing one base configuration across folders, or when a configuration is rejected, a keyword list will not import, or a word is wrongly listed in the glossary.
Use when writing or improving a definition in a coinagerie glossary, filling a bare `- **word**:` bullet or an empty JSON value, reviewing a definition that restates its word or reads like a dictionary entry, or explaining a word whose meaning in this codebase differs from its everyday English meaning.
Use when coinagerie --check exits non-zero, a run names missing or stale words, refuses a glossary it cannot parse, or rejects a configuration or command line, when --write deleted a definition, when a glossary headword looks wrong or surprising, when a check passes locally and fails in CI, or before adding a word to exclude.words, narrowing include, or deleting a glossary block to make a failing check pass.
Install, set up and use the tailwindcss-concentric plugin to fix and prevent mismatched nested corner radii in Tailwind CSS v4 projects. Use when the user asks to add or set up tailwindcss-concentric, to audit or fix nested border radii, or when building or reviewing UI where a rounded element sits inside a padded or bordered rounded parent (cards with media, buttons in search bars or toolbars, segmented controls, selection rings, modals, input groups), when you see hard-coded inner radii such as rounded-[calc(...)], or when the user mentions concentric, nested, inner or outer corner radius
Contextualizes the business of an existing system (brownfield) for AI-first work: builds the capability map (what the system does today), surfaces business rules hidden in code, and creates the business ↔ code glossary with the ARCH, all with evidence, then pauses for human approval. Use when the BA role starts on an existing system without a PRD or backlog, or when asked to document what a legacy system does, extract its business rules, or build a domain glossary
Contextualizes an existing repository (brownfield) for AI-first work on it: verifies how to start and test it, creates the CodeGraph index, reconstructs the actual layer map, documents inherited decisions as ADRs, assembles the pattern catalog (canon vs. legacy), the debt map and verifiable build rules, all with file:line evidence, then pauses for human approval. Use when the ARCH role starts on an existing codebase without a PRD, or when asked to contextualize, onboard, document, map or understand a legacy/brownfield repository before working on it
Find the mistakes agents keep repeating in this repo and make each one impossible. Try architecture first, then types, then a lint whose error names the fix, then a test, and write docs last. Prove each check fails on a real past mistake. Repeat this each time the operator corrects you. Use for /correct.
End-to-end Course & Class Management coordinator. Guides teachers, lecturers, and instructors from RPS/RPP analysis, teaching module creation, tiered exercises, 16:9 slides, web and AR gamification, automated assignment assessment, OBE dashboards, portfolio reporting, through to textbook connections.
Design and create interactive, web-based educational mini-games (HTML5 Canvas/JS/CSS) for each lesson. Turn teaching topics into concept-sorting games, logic puzzles, quick-guess arenas, branching scenario simulations, or crosswords playable in desktop or mobile browsers.
Install the Datadog Agent on a macOS machine with the macOS installer, with infrastructure metrics, logs, and APM trace intake on by default, plus optional live processes, APM instrumentation, custom tags, and service management. Use when the user wants to install the Datadog Agent on their Mac, set up Datadog on macOS, or monitor a macOS developer or CI machine with Datadog. Does not install the Agent on Linux, Windows, Docker, or Kubernetes (use dd-agent-install-linux, dd-agent-install-windows, dd-agent-install-docker, or dd-agent-install-kubernetes) and does not target managed cloud fleets. Requires a Datadog account with a valid API key and application key; run `dd-account-setup` first if they are missing.
Install the Datadog Agent on Windows Server hosts—a single machine or a fleet—with support for cloud rollouts (AWS Systems Manager, GCP VM Manager, Azure Run Command), MSI properties, FIPS, custom tags, and advanced configuration. Infrastructure metrics, logs, and APM trace intake are enabled by default. Use when the user wants to install the Datadog Agent on a Windows server, deploy Datadog to Windows VMs, or set up the Datadog Agent on Windows Server 2022 or another Windows Server version. Does not install the Agent on Linux, macOS, Docker, or Kubernetes (use dd-agent-install-linux, dd-agent-install-macos, dd-agent-install-docker, or dd-agent-install-kubernetes) and does not connect a cloud provider integration. Requires a Datadog account with a valid API key and application key; if they are missing, run `dd-account-setup` first on commercial sites, or supply them directly on government sites.
Set up Datadog Storage Management for AWS S3 by configuring S3 Inventory on the buckets selected by the user and provisioning destination buckets, a destination bucket policy, a Datadog read policy, and Storage Management sync with the DataDog/storage-management-datadog/aws Terraform module. Supports multi-Region and SSE-KMS destinations and optional S3 access logs. Use when the user wants to set up Storage Management for S3 buckets, monitor S3 storage in Datadog, or enable S3 storage management. Does not connect a new AWS account to Datadog or deploy the Datadog Forwarder Lambda (that is `dd-aws-log-forwarding`). Requires an existing role-based Datadog AWS integration; run `dd-aws-integration` first if it is missing.
Set up Datadog Database Monitoring (DBM) for PostgreSQL on AWS RDS - enable pg_stat_statements on the instance's parameter group, then deploy a managed Agent with CloudFormation, provision one with Terraform, or configure an existing Agent, plus the read-only datadog monitoring user. Use when the user wants to set up Database Monitoring for an RDS Postgres database, monitor query performance on a Postgres RDS instance, enable DBM for Postgres on RDS, or use Terraform to set up DBM for an RDS Postgres database. Does not handle Aurora, RDS Multi-AZ DB clusters, or read replicas (point the user at the source writer or the DBM docs instead), and does not set up DBM for other engines. Requires the Datadog AWS integration for the AWS account that holds the instance; run `dd-aws-integration` first if it is missing.
Set up Datadog Cloud Cost Management for GCP - enables the BigQuery and BigQuery Data Transfer APIs, walks the user through the Cloud Billing detailed-usage export to BigQuery, then uses Terraform to provision the GCS bucket and the predefined IAM role bindings Datadog documents for the Datadog service account, and registers the cost config through datadog_gcp_uc_config. Use when the user wants to track their GCP cloud costs, asks to set up a GCP billing export for Datadog, or asks to set up Cloud Cost Management for GCP. Does not create the GCP integration or the Datadog service account itself. Requires an existing Datadog GCP integration; run `dd-gcp-integration` first if it is missing.
Set up GCP log forwarding to Datadog with Terraform. Creates a Dataflow pipeline running Google's Cloud_PubSub_to_Datadog template, fed by log sinks on the selected projects and/or folders through a Pub/Sub topic and subscription, with a dead-letter queue, a GCS staging bucket, and Secret Manager holding a dedicated Datadog API key. Use when the user wants to forward GCP logs to Datadog, set up GCP log forwarding, or send Cloud Logging to Datadog. Does not create the GCP integration itself or set up metrics collection. Requires the Datadog GCP integration; run `dd-gcp-integration` first if it is missing.
Debugs issues users encounter with Tuist-generated projects by reproducing the scenario locally, building Tuist from source when needed, and triaging whether it is a bug, misconfiguration, or something that needs team input. Use when users report generation failures, build errors after generation, or unexpected project behavior.
Builds a one-screen executive report dashboard for teachers and lecturers. Presents all key class indicators (assignment submission rate, average CPMK, grade distribution, alerts for students needing attention) with a responsive Mobile-View First design and strict credential security.
Connect a JavaScript or TypeScript app to Fallow Cloud production runtime coverage. Use when asked to set up Fallow Cloud, add the Fallow beacon, send runtime coverage to Fallow Cloud, or find out why Fallow Cloud shows no data for a repository. Do not use for local code-quality tooling; use fallow-setup for that. Do not use to read runtime data of a repository that is already set up; use the fallow skill for that.
Use when the user asks to fix, patch or upgrade a Node.js vulnerability (e.g. 'Please fix my Node vulnerability'), the October 21, 2026 Node.js security release, CVE-2026-111497 (Brotli RCE), or end-of-life Node 20, across Docker images, .nvmrc/.tool-versions, setup-node CI, Lambda and IaC. Read-only inventory, approval gate, regression tests, focused PRs.
MANDATORY - use before planning, building, or redesigning an ArcGIS application, including requests that do not specify an ArcGIS technology. For existing applications, use when a change could affect architecture, product, technology, data access, or authentication. Review context, clarify consequential unknowns, and route to relevant authentication and build skills.
Guides day-to-day work in Tuist-generated Xcode workspaces, including generation, build and test commands, and buildable folders. Use when working in a Tuist-generated project or when users mention `tuist generate`, `xcodebuild test`, or generated workspaces.
Design and build self-contained interactive HTML5+CSS3+JS presentation slides in 16:9 format, with keyboard navigation, full-screen mode, interactive QR codes for students, and print-ready layouts for instant PDF export without distortion.
Aggregate and read real user monitoring (RUM) data via KubeSense MCP — sessions, views, actions, errors, resources and long tasks from browser and mobile apps. Covers the rum_<type> signals, the per-event-type field catalog, which fields accept LIKE or group-by, nanosecond durations and their _ms/_s twins, feature flags and attributes, and the dashboard panel shape. Counts, rates and percentiles with analyze-telemetry, raw rows with search-rum. There is no alerting, SQL or SPL for RUM.
Diagnose why a Mac (especially one running Claude Code, other AI coding tools, or several Electron-based IDEs at once) is slow, hanging, or freezing due to memory pressure and swap thrashing. Covers reading real swap/memory numbers, finding MCP-server and plugin process bloat, spotting orphaned or duplicate background processes, safely clearing regenerable caches, and recognizing when a killed process keeps respawning because something else supervises it. Use this whenever the user says their Mac/MacBook is slow, laggy, hanging, freezing, "using too much memory," has "high memory pressure," or asks to find/stop/kill whatever is eating RAM or CPU. Every fix in this skill is reversible (stop, kill, disable, clear cache) — it never deletes or uninstalls anything, and it says so explicitly before running anything.
Rewrites a bolted-on feature so it reads as if it was always part of the original design. Use when a diff, branch, or commit adds a feature through special cases, flags, aliases, parallel components, or preservation tests instead of reshaping the existing code
Delegate a coding task to the MiniMax Code CLI (`mcode`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to MiniMax Code or a MiniMax model—phrasings like "have MiniMax do X", "delegate this to mcode", "run it through MiniMax Code", or "use MiniMax to implement/fix/refactor"—or to run a queue of coding tasks through MiniMax Code while staying the reviewer. DO NOT USE for tasks small enough to do inline, or when the user wants the code written directly without delegating
Builds a professional Excel spreadsheet workbook (.xlsx) for calculating Outcome-Based Education (OBE) assessment fulfillment, mapping CPL and CPMK attainment per student, grade distribution matrices, and visualizing accreditation achievement charts.
Operate Orca-managed worktrees, folder contexts, terminals, repos, automations, artifacts, skill sharing, worktree comments, and Orca's embedded browser through the `orca` CLI. Use when the user says "$orca-cli", "Orca worktree", "child worktree", "spawn codex/claude in a worktree", "read/wait/send Orca terminal", "handoff" / "handover" / "give this to another agent", "Orca browser", "orca artifacts", or "share skills". Prefer it over raw git worktree, ad hoc PTYs, or Computer Use when Orca state is involved. Use Computer Use only when a visible window needs GUI control that a CLI, filesystem, or API cannot do.
Read org state and change Setup through the Headless 360 (H360) MCP tools discover, describe, dispatch_readonly, and dispatch. Use when H360 tools are present and the answer is org state no skill or sf command reads: Setup settings, feature or license access, org-specific caps, health checks, release updates. Also H360 setup and errors (401, 403, 404, ROUTE_NOT_FOUND, INVALID_ARGUMENT). DO NOT TRIGGER for records, SOQL, schema, limits in sf org list limits, deploys, or which-plugin questions.
Transform mocked-up UI specs (specs-ui skill project) into a purely UI clickable prototype in HTML + Tailwind + Alpine.js: main flows are wired up (navigation, forms, states, simulated data stored in the browser, reset), with guided test scenarios for user testing and demos, published as a multi-page artifact. Use whenever someone wants a prototype, clickable mockup, interactive demo, user-testable flow, user test, or a version "you can click through" based on existing specs or mockups, even without saying "prototype".
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.