Skills
101,174 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Analyze staged/unstaged changes and create semantic conventional commits with context about WHY, not just WHAT. Auto-detects commit type and scope from the diff. Supports optional type/scope arguments. Usage - /commit-smart, /commit-smart fix, /commit-smart refactor api
Guidance for Microsoft Entra Conditional Access (CA) and multifactor authentication — the Zero Trust policy engine that enforces grant/block decisions based on user, device, location, app, and risk signals. Covers a baseline 6-policy set, authentication strengths for phishing-resistant MFA, report-only rollout, break-glass exclusions, session controls, and policy lifecycle. WHEN: Conditional Access, MFA enforcement, require compliant device, block legacy authentication, Zero Trust access policy, phishing-resistant MFA, FIDO2, passkey, session controls, sign-in frequency, CA policy design, require MFA for admins, baseline CA policies, report-only mode, What If tool, exclude break-glass, named locations, app protection policy grant, terms of use. DO NOT USE for risk-based detection signal sources (use entra-id-protection) or PIM activation (use azure-pim).
Create, schedule, duplicate, inspect, cancel, archive, and troubleshoot Vapi outbound Campaigns. Use for persistent multi-contact calling, CSV or API contact personalization, campaign concurrency, campaign webhooks, pre-dial eligibility, contact outcomes, or rerunning an audience. Do not use for a single call or a simple one-off /call batch.
Deploy, configure, and operate CrowdSec Security Engine, cscli, remediation components, acquisition pipelines, and AppSec WAF. Use for Linux or Docker installation, detection-to-blocking design, incident review, and safe changes. Do not use for generic firewall, Kubernetes, or reverse-proxy design; route those to the named platform skill and use this skill for CrowdSec integration.
Copy, clone, duplicate, rebind, and execute Gen1 dataflow save-as upgrade operations via CLI (az rest / curl) against Power BI REST and Fabric REST APIs. Covers Gen1 to Gen2.1 upgrade save-as, cross-workspace Gen1-to-Gen2.1 copy flows, connection rebinding, output changes, readiness snapshots, and SaveAsNativeArtifact execution. Use when the user wants to: (1) discover Gen1 dataflows, (2) assess save-as readiness, (3) upgrade Gen1 into Gen2.1, (4) create a Gen2.1 copy from a Gen1 dataflow in another workspace, (5) rebind connections or validate saved data. For creating/editing Gen2 dataflows from scratch, previewing candidate M, binding ordinary authoring connections, or copying Gen2 dataflows, use `dataflows-authoring-cli`. Triggers: "save Gen1 dataflow", "upgrade dataflow", "Gen1 Gen2 readiness", "Gen1 Gen2 save-as", "saveAsNativeArtifact", "copy Gen1 dataflow", "duplicate Gen1 dataflow", "clone Gen1 dataflow", "rebind Gen1 dataflow connections".
Product-interface quality gate for evidence-based UI decisions. Use when designing or implementing a user-facing UI, reviewing its usability, refining an existing interface, or changing responsive, stateful, overlay, or cross-platform behavior.
Visual design principles from Design for Hackers — the visual + surface core of a design-foundations system. Use when building or improving UI/frontend visual design: choosing fonts, building color systems, establishing a unique design direction or design DNA, designing for a device class (phone, TV, watch, in-car, kiosk, voice), auditing designs, choosing animation/motion/3D libraries, or polishing before shipping. Also when something 'looks wrong' or 'feels off', or a design looks generic or AI-generated. Triggers on: pick fonts, color palette, design audit, visual hierarchy, redesign, unique design, 10-foot UI, device layout, add motion, which animation library, GSAP, three.js. Not for: graphic-design tools (Figma/Sketch), print, logos, illustration; or sibling-pillar concerns — operability (usability), UX writing (content-design), chart/data encoding (data-viz), dark patterns (deceptive-patterns), user flows and journey maps (journey), design tokens (design-systems), agent/LLM interfaces (ai-native).
Design movements from Arts and Crafts through Bauhaus, Swiss Style, Memphis, and grunge to flat design and current styles, with context, visual markers, and UI tokens for each. Use when choosing an aesthetic direction, explaining why a style reads the way it does, or applying a period style without misusing it.
Use when the user says 'build a design system for my site', 'extract design tokens', 'capture my brand', or 'build my style guide', or after a rebrand. Reads representative pages, theme files, and media to extract logo, colors, typography, spacing, and components, then writes a visible style-guide page.
Produces this framework's ADR and system-design artifacts from bundled templates with trade-off analysis. Use when designing a system or component, choosing between architectures, or recording an architecture decision in the planning flow.
Deep-clean AI-generated code across a whole codebase — dedupe, dead code, circular deps, weak types, defensive try/catch, legacy paths, AI-slop comments. JS/TS, Ruby/Rails, Go. Run after heavy AI sessions or before merging long-running branches. Whole codebase (vs `/simplify` for recent edits).
Start here for Dessn (create.dessn.ai), a design canvas where every frame is the person’s real app running live from their own code. Connected users ask in their own words, mostly without naming Dessn (design a new settings page, make variations of a page, a page I can click through, review our onboarding and propose changes, map our sign-up flow, show the checkout flow with each step and where errors go), and what they ask to see goes on their Dessn canvas as live frames they can click through and share, while a local file or written answer stays on this computer alone. Also use when they mention Dessn, paste a create.dessn.ai link, lay out a board, review frames, or bring a Dessn design into their code. Covers connecting and signing in, finding projects and pages (including when none exist yet, setting up their own app from the current folder, or using a starter app whose frames are interactive when there is no app), which Dessn tool fits each job, and which Dessn skill to load next
Lay out Dessn canvas pages so a person can read them without the chat. Create QA boards of an app’s pages grouped by area; map user flows of live app pages joined by labelled arrows (the main path in a row, each failure shown as a real state of its step’s page beneath it); mark what’s broken where it happens when auditing a flow; draw diagrams from Mermaid or nodes and edges; add title and caption notes, sections, and frames fitted to their content. Use for “put every surface of our app on a page”, “map our sign-up flow”, “show the checkout flow with each step and where errors go”, “lay out the onboarding journey for a new user”, “audit our checkout flow and flag what’s broken”, “draw this architecture or state diagram on the canvas”, “tidy / arrange this board”, “label these frames”, or “make the frames fit”
Work with the app’s code in Dessn’s cloud when there is no local checkout. Find a surface from the person’s words, find the code behind it, explain how it’s built, change it on a copy, look at the result, put it on a board beside the original, and undo—all through the Dessn MCP. Use when the user has no code locally (a PM or designer with Claude Desktop, claude.ai or Claude Code in an empty folder) and wants to understand, change or explore a surface of their Dessn project, such as “where does this text come from”, “change the copy on the pricing page”, “how is the empty state built”, “make this friendlier and show me”
Design on the person's Dessn canvas, as Dessn's own design agent does. People ask in their own words, mostly without naming Dessn (design a new settings page, make 3 pricing variations, add an empty state to the projects list, make a landing page for a feature, a pricing page I can click through, review our onboarding and propose changes, explore different directions, mock this change, show it at phone width). With Dessn connected, each request becomes frames of their real app on their canvas—a copy whose code you write in seconds with no credits, which they can click through, compare, and share, kept 1:1 with their local code by git; a local edit appears only on this computer. Reuse the app's components (dessn_designSystem_get), never recreate them; check each result against a real picture; mock a UI change as a copy of the app's frame whose code you write; and use HTML frames for notes and annotations. Also for a design-system page in the app, denser or darker versions, and empty, error, and loading states.
Move code between the person's local clone of their app and Dessn with git, in both directions, through the Dessn MCP. Push the local checkout to a Dessn frame (their own copy) and put frames of its routes on a board; or bring a Dessn frame's changes home to the clone as git (status, a patch to git apply, its commits for git am, a bundle to git fetch) and build it. Use when the user codes locally in a clone of a Dessn project's repository, such as “push my local changes to Dessn”, “put my branch on the board”, “sync my working tree to Dessn”, “take this frame's change into my repo”, “bring the design from Dessn into my code”, or “apply this frame”.
When the user wants to create email sequences for developers including onboarding, product updates, re-engagement, or changelog communications. Trigger phrases include "developer emails," "onboarding sequence," "email drip," "developer newsletter," "changelog email," "re-engagement campaign," or "email cadence."
This skill should be used for personal operating-system workflows: content creation, voice consistency, relationship lookup, meeting preparation, weekly review, goal tracking, personal brand management, and network management.
Design, build, evaluate, govern, monitor, evolve, and retire digital twins and federated twin universes for software systems, engineering processes, agentic software factories, infrastructure, and cyber-physical operations. Use when a task involves digital-twin architecture, digital thread, simulation, predictive maintenance, twin health, agent authority, or dark-factory design. Do not use for ordinary observability dashboards, static dependency graphs, generic AI governance, or operating one named infrastructure tool without a twin-specific representation and feedback loop.
Provides comprehensive guidance for Docker Compose including multi-container applications, service definition, networking, and volumes. Use when the user asks about Docker Compose, needs to orchestrate multiple containers, define docker-compose services, or manage multi-container applications.
Search auto-generated codebase documentation for function signatures, API docs, class definitions, and code comments. Use when the user asks to "search docs", "find documentation", "look up a function", "check the API", or before implementing changes to verify correct signatures and patterns.
Systematically explore and test a web application to find bugs, UX issues, and other problems. Use when asked to "dogfood", "QA", "exploratory test", "find issues", "bug hunt", "test this app/site/platform", or review the quality of a web application. Produces a structured report with full reproduction evidence -- step-by-step screenshots, repro videos, and detailed repro steps for every issue -- so findings can be handed directly to the responsible teams.
Edge computing with Cloudflare Workers, Deno Deploy, Bun, Vercel Edge Functions, AWS Lambda@Edge, and edge databases (Turso, D1, DynamoDB Global Tables). Use when building low-latency edge applications, edge-side rendering, or globally distributed compute.
Portable Effect patterns for robust promise execution. Use when wrapping async operations with timeouts, retries, tagged errors, caching, concurrency, pattern matching, or tracing - all designed to resolve to a plain Promise via Effect.runPromise.
Expert code review for embedded/firmware projects with dual-model cross-review (Claude + Codex via ACP). Detects memory safety, interrupt hazards, RTOS pitfalls, hardware interface bugs, and C/C++ anti-patterns. Trigger when user asks to review embedded/firmware/MCU code changes, diffs, or PRs. Examples: 'review firmware-pro2 的改动', 'review the NFC changes', '/embedded-review ~/path/to/repo', '/embedded-review ~/path/to/repo HEAD~5..HEAD', '/embedded-review <github-pr-url>'. Target environments: bare-metal MCU (STM32, nRF, ESP32, RP2040), RTOS (FreeRTOS/Zephyr/ThreadX), Linux embedded, mixed C/C++ firmware.
List, inspect, and monitor Fabric Eventstream real-time ingestion pipelines via the Items REST API. Discover Eventstreams across workspaces, decode base64 graph topologies tracing event flow from source through operators to destination nodes. Validate connection IDs, wiring, retention policies (1-90 days), and throughput levels. Retrieve Custom Endpoint Kafka credentials via Topology API. **Invoke this skill** to: (1) list Eventstreams, (2) inspect Eventstream topology showing sources and destinations, (3) validate Eventstream configurations, (4) check Eventstream retention policy and throughput level, (5) get connection strings. Triggers: "list eventstreams", "inspect eventstream", "describe eventstream topology", "eventstream operator nodes", "eventstream sources and destinations", "eventstream health", "eventstream status", "eventstream configuration", "eventstream retention", "eventstream throughput level", "eventstream connection string", "custom endpoint credentials", "check eventstream".
Create and populate Feishu (Lark) Bitable (multidimensional tables) with automated cleanup. Use when the user needs to: 1. Create a new Bitable from scratch with clean structure (no placeholder rows/columns) 2. Batch create fields and records in a Bitable 3. Convert structured data into a Bitable format 4. Create data tables for research, comparison, or tracking purposes Automatically handles: empty placeholder row cleanup, default column removal, intelligent primary field naming, and batch record creation.
Launch and trade Agent-ID-gated tokens on FLAIP (https://flaip.xyz, BNB Chain). Use when the user wants to create or launch a token on FLAIP / AGENT//PAD, buy or sell a FLAIP token by contract address (CA) or pool, claim graduated tokens, withdraw or refund, or check a FLAIP transaction. Builds unsigned transactions through the FLAIP API and signs them with the Binance Agentic Wallet contract-call after one user confirmation.
Comprehensive guide for building 2D games with Flutter and Flame Engine. Covers game architecture, sprites, animations, collision detection, input handling (keyboard/joystick/gamepad), camera systems, Tiled maps, audio, particles, and UI overlays. Use when creating or refactoring Flutter games.
Analyze, design, critique, balance, and rework video game mechanics, systems, and interface, grounded in a cross-game pattern catalog and design frameworks (MDA, core loops, flow, game feel, economy). The first stop for any design-touching change or discussion, including mid-implementation in an engine like Godot. Use when designing, balancing, brainstorming, or reworking a mechanic, progression, economy, loadout, inventory, loot, shop, enemy or difficulty scaling, reward, menu, or UI system; when tuning drop rates, prices, curves, or scaling; when asking whether something is fun, too grindy, or has a dominant strategy; when designing menus, HUD, transitions, onboarding, dialogue, guidance, input prompts, or accessibility; when capturing, pitching, or iterating a game idea, concept, vision, design pillars, or GDD, persisted as a living concept bundle; or when self-authoring game systems while building one. Rides alongside godot, which writes the engine code, while this decides what that code should do.
Expertise in real-time multiplayer networking, lag compensation, and authoritative server architectureUse when "multiplayer game, netcode, client-server game, P2P networking, lag compensation, rollback networking, game server, matchmaking, lobby system, player synchronization, state replication, dedicated server, authoritative server, tick rate, network prediction, interpolation, extrapolation, GGPO, lockstep, real-time multiplayer, networking, multiplayer, gamedev, realtime, client-server, p2p, netcode, synchronization, lag-compensation, matchmaking" mentioned.
gate-dex CLI swap skill: one-shot DEX swap (quote → GV checkin → sign → broadcast) via the swap command. GV checkin built in — no external binary. Supports EVM multi-chain and Solana, with ERC20 approve flow. Use this skill whenever the user wants to swap, exchange, buy, sell, convert, or bridge tokens via the gate-dex CLI. Trigger phrases include swap, exchange, buy, sell, convert, bridge, cross-chain, quote, swap history, swap detail, swap-tokens, bridge-tokens.
Use when implementing OCI GenAI inference APIs, troubleshooting rate limits or token errors, optimizing GenAI costs, or handling sensitive data (PHI/PII) in prompts. Covers model selection, cost calculations, token management, response validation, and healthcare/compliance considerations.
Generate a project's Technical Design Document + ADRs (MADR format) + acceptance criteria, interviewing the user decision by decision starting from their PRD (and optionally their Design.md or an existing repo). Use when the user asks to generate, draft, or create a technical design document, system architecture, or architecture decision records (ADRs) for a project, greenfield or brownfield.
Configure GitLab project security scan profiles with glab. Use when enabling or disabling SAST, dependency scanning, secret detection, dependency-scanning auto-remediation, business_logic, skillguard, triage_and_remediation presets, or custom profile IDs; or checking scan-type status, including container_scanning. Triggers on GitLab security config, security scan profile, enable SAST, dependency scanning, secret detection, container scanning status, business logic, SkillGuard, triage and remediation, glab security
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.