Skills
98,873 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Email marketing automation for HubSpot/Klaviyo/Mailchimp (and similar ESPs): design trigger-based lifecycle workflows (welcome, onboarding, nurture, win-back, cart abandonment), define segmentation + suppression/frequency policies, troubleshoot deliverability (SPF/DKIM/DMARC, one-click unsubscribe, list hygiene), and measure incrementality/ROI (holdouts, RPE, retention economics).
Guide for accessing financial and economic data in MATLAB using the Datafeed Toolbox. Covers Bloomberg (market data via bloomberg/blp/bloombergHypermedia), FRED (Federal Reserve economic data via fredrs), Haver Analytics (economic data via haver/haverdirect/haverview), and LSEG Datastream (historical data via datastreamws). Use when connecting to any of these data providers from MATLAB.
Build industrial-grade SCADA/HMI dashboards in MATLAB following industrial-HMI conventions (ISA-101-aligned): gray-field philosophy, alarms at source, write safeguards, fixed-range trends, drill-down layout. Produces a real App Designer app (.mlapp, or plain-text .m+.xml on R2026b+) by handing serialization to the matlab-build-app skill when available, and falls back to a programmatic .m app otherwise. Use when wrapping OPC UA / Modbus / MQTT / OSI PI / PI AF monitoring scripts into a live operator app, building plant overviews, designing operator dashboards, or any time a user asks for a "SCADA dashboard", "HMI", "plant dashboard", "operator screen", or "industrial monitoring app" in MATLAB. Trigger on: SCADA, HMI, industrial dashboard, plant overview, operator screen, uigauge, uilamp, alarm banner, gray-field, ISA-101, OPC UA dashboard, setpoint, write safeguards, alarm visualization, OSIsoft PI, AVEVA PI, PI Server, PI Data Archive, PI AF, PI Asset Framework, piclient, afclient.
MatrixScan Batch (MatrixScan, BarcodeBatch, legacy BarcodeTracking) in web/browser (TypeScript/JavaScript) projects (@scandit/web-datacapture-barcode) — tracking and scanning multiple barcodes at once. Use for integration, settings and symbologies, per-barcode brushes, HTML-element AR overlays, manual feedback, lifecycle, SDK version migration, or troubleshooting.
Guide for creating custom .NET MAUI handlers, customizing existing handlers with property mappers, and implementing platform-specific native views. Covers PrependToMapping/ModifyMapping/AppendToMapping, PropertyMapper, CommandMapper, partial handler classes, and handler registration. USE FOR: "custom handler", "PropertyMapper", "AppendToMapping", "PrependToMapping", "ModifyMapping", "CommandMapper", "platform-specific rendering", "native view", "handler registration", "custom control renderer". DO NOT USE FOR: platform API calls without custom controls (use maui-platform-invoke), data binding (use maui-data-binding), or gesture handling (use maui-gestures).
Max — the GTM-system copilot: the advisor that sits ABOVE Victor Shulga's skill stack and guides a person who downloaded the skill bundles. Max onboards them, diagnoses where their agency is, routes them to the RIGHT skill for their stage, explains what each skill is for and when to run it, enforces the gates, and always hands off to one concrete next action. Use whenever someone asks: "which skill do I use", "where do I start", "how does this system work", "what should I do next", "explain [skill]", "I have a GTM/outbound/sales/ retention problem — help me", "я загубився в скілах", "з чого почати", "що робити далі", "поясни скіл X", "I have no SDR", "show me the team", «у нас нема SDR», «покажи команду», «Макс, тренажер: <компетенція>», or opens the bundle for the first time. Max ROUTES and COACHES; it sends you to the skill that does the task (the trainer drill is the one it runs). NOT for running a specific skill end-to-end (invoke that skill directly) and NOT for inventing GTM advice outside the system.
Manage Workers/KV/R2/D1/Hyperdrive via Cloudflare MCP, perform observability/build troubleshooting/audit/container sandbox operations. Triggers: worker/KV/R2/D1/logs/build/deploy/screenshot/audit/sandbox. Three permission tiers: Diagnose (read-only), Change (write requires confirmation), Super Admin (isolated environment). Write operations must follow read-first, user confirmation, post-execution verification.
A security testing methodology for the MCP (Model Context Protocol) protocol. Triggers when the target environment uses MCP Servers to integrate external tools, when the security of MCP tool descriptions needs to be assessed, or when testing the security boundaries of Agents invoking tools via MCP. Covers: tool description poisoning, rug pull attacks (dynamic tampering), instruction override (Shadow Tool), hidden instructions (ANSI/Unicode), cross-Server attacks, token theft, schema manipulation, and context overflow.
Video/audio/image processing with FFmpeg and ImageMagick. Tools: FFmpeg (video/audio), ImageMagick (images). Capabilities: format conversion, encoding (H.264/H.265/VP9/AV1), streaming (HLS/DASH), filters, effects, thumbnails, watermarks, batch processing, hardware acceleration (NVENC/QSV). Actions: convert, encode, resize, crop, compress, extract, merge, stream, transcode media. Keywords: FFmpeg, ImageMagick, video encoding, audio extraction, image resize, thumbnail, watermark, HLS, DASH, H.264, H.265, VP9, AV1, codec, bitrate, framerate, resolution, aspect ratio, filter, overlay, concat, trim, fade, batch processing. Use when: converting video/audio formats, encoding with specific codecs, generating thumbnails, creating streaming manifests, extracting audio from video, batch processing images, adding watermarks, optimizing file sizes.
AI-first security scanning with Medusa. 3,000+ detection patterns covering AI/ML, agents, MCP, RAG, prompt injection, and traditional SAST vulnerabilities. Wraps Medusa CLI with SARIF/JSON parsing, structured finding output, OWASP mapping, and remediation guidance.
Use when the user asks to "remember project context", review saved findings, initialize runtime memory, archive stale work, reconcile notes, or erase a subject; manages authorized HOT/WARM/COLD working memory across all disciplines while preserving registry event ownership and privacy controls. Not for changing canonical registry facts - route those through the owning registry. 项目记忆/跨会话
Dedicated Skill for the doors, windows and window sealing pitfall guide. Covers the full process including window selection, entry doors and interior doors, window sealing and floor-to-ceiling windows, curtains and shading, installation sealing and hardware, and door-and-window selection overviews, integrating the highlights of the Xi'an Renovation Classroom members' edition knowledge base in reverse order, helping users renovate with less hassle and fewer pitfalls in doors, windows and window sealing.
Generate an /insights-style HTML report across MANY of your agent coding sessions (Claude Code and OpenAI Codex): what you work on, how you use the agent, your wins, your friction points, and concrete fixes to become more effective — precise goal-setting, not trusting output blindly, scoping, context hygiene, right-sizing models. Reads local session history and writes one self-contained report.html. Use when the user asks to "разбери мои сессии", "оцени как я работаю с агентом", "мой insights", "session insights report", "review how I work", "на что обратить внимание", or runs "/mentor", "/mentor claude", "/mentor codex", "/mentor both". Do NOT use for reviewing the CODE produced (that's code-review) or building the agent.
MKD (`mkd` — Make Decision) — a local, browser-based decision-deck CLI the AI drives to collect structured human feedback: decisions with justified options, questions, report reactions and answerable tables, one screen at a time (Slack Catch-Up style). WHEN to use: the AI has MORE THAN ~3 decision points or a long multi-section report/plan that needs the user's verdict point by point. It beats `AskUserQuestion` (capped at ~4 questions × ~4 options, terminal-bound, no rich free-text, cannot show reference content) and beats an inline prose questionnaire (unanchored replies the AI must guess-map back). HOW it works: the AI writes a spec JSON of `items` (`decision` | `question` | `report` | `table`), runs `bun <skill-dir>/cli/index.ts <specPath>` (no install, no binary), and the CLI renders a self-contained deck page under `~/.mkd/`, opens the browser and EXITS immediately — the user answers at their own pace (progress rail, arrow keys, skip = decide later, live stats, localStorage persistence) and pastes the copied Result JSON back into the chat, which the AI treats as the execution contract. Inside Orca (multi-agent IDE) the default changes: the CLI serves the deck, opens it in a browser tab bound to the current worktree and BLOCKS until the user presses Send, then prints the Result JSON on stdout, so the AI must run it in the background (`--copy` restores the copy-paste flow). Add `--wait` outside Orca only when the answer is needed the same turn (blocking loopback server, Result JSON on stdout). In Claude Code with the `Artifact` tool, pass `--artifact`: the CLI renders a claude.ai Artifact page, the agent publishes it with `capabilities: {db: {}, comments: {}}` and arms a watch, and the user's Send to Claude wakes the session with the Result in the artifact's db (no localhost, no copy-paste); other harnesses keep the flows above. HARD RULE: every decision option MUST carry a written justification (value + cost), and the recommended option's justification must state WHY it is recommended — the validator rejects options without one. Triggers on: `mkd`, `make decision`, `decision deck`, `catch-up`, `catchup`, `decision form`, `feedback UI`, `ask me point by point`, `let me answer in the browser`, `review this plan point by point`, `react to each section`, `more than three questions`, `decisiones de auditoría`, `tomar decisiones`, `mkd artifact`, `answer in a claude.ai artifact`. Use this skill even when the user does not say `mkd` — if the task is to collect granular decisions or anchored feedback on many points, this is the right tool. Do NOT use for: a trivial 1-2 option pick or a single yes/no (use `AskUserQuestion`), or any non-interactive / CI / one-shot output where there is no human at a browser.
Bootstrap or maintain MkDocs Material docs sites with GitHub Pages, strict two-pass i18n plus default-language llms.txt, copy-to-LLM, page/nav helpers, and zh-TW translation. Use when setting up or publishing docs, adding bilingual or multilingual docs, repairing nearly empty llms.txt or i18n strict-build failures, or migrating an older mkdocs-site-bootstrap site. Consent-gated; never rewrites existing docs without approval.
Manage project learnings — small, atomic, validated patterns that the agent should remember across all skills and sessions. Different from /money-save (which captures full session state); learnings are individual insights that get auto-loaded into every other money-* skill's context. Use when the user has just discovered something worth remembering — a customer pattern, a pricing insight, a channel that works, a failure mode. Triggered by: 'remember this', 'log a learning', 'this is a pattern', 'show learnings', 'what have we learned', '记住这个', '存入经验', '查看经验库'.
Run all four review skills (investor, customer, operator, skeptic) on a business plan in sequence, then auto-decide where verdicts agree and surface only disagreements + taste decisions to the user. One command = full review gauntlet, intelligently filtered. Use when a plan is ready to be stress-tested before commit. Triggered by: 'panel review', 'review panel', 'run all reviews', 'review gauntlet', '审议会', '四方评审'.
Weekly business retrospective. Reads all snapshots, learnings, skill-usage telemetry, and (if provided) revenue data from the past week. Outputs: what got decided, what shipped, what's stalled, skill activity, suggested next focus. Surfaces unused skills as activation prompts. Use at end of week or end of sprint. Triggered by: 'weekly retro', 'business retro', 'what did we ship', 'how's the week going', '周复盘', '业务回顾', '本周复盘'.
Review a business plan from the perspective of the named target customer. Asks: would they actually pay this price for this solution today? Outputs a verdict with one of four modes — PAY NOW / PAY WITH FRICTION / WOULDN'T PAY / WRONG ICP. Use when the user has a defined ICP and pricing and wants to stress-test whether the value prop survives contact with reality. Triggered by: 'customer review', 'would they pay', 'pricing reality check', '客户视角', '会付费吗'.
Review a business plan or product strategy through the eyes of a smart-money seed/Series-A investor. Asks the brutal questions: Is this fundable? What's the moat? What's the exit story? Outputs a verdict with one of four modes — SEED VIABLE / LATER ROUND ONLY / BOOTSTRAP-ONLY / UNFUNDABLE. Use when the user has a business plan and wants to know if it would survive an investor pitch. Triggered by: 'investor review', 'is this fundable', 'VC perspective', '投资人视角'.
Save the current business state to disk so future Claude Code sessions can pick up where you left off. Captures confirmed conclusions, ruled-out directions, open hypotheses, and the next recommended skill. Use when the user has just reached a conclusion in /money-discover, /money-strategy, /money-diagnose, or any pipeline phase, and wants to lock the state in. Also triggered by: 'save this', 'checkpoint', 'remember this', 'lock it in', '保存', '存档', '记下来', '这个结论留着'.
Codify a successful workflow from the current conversation into a permanent project-local skill. After you discover (e.g.) a cold-email sequence that converts, an ad-creative format that beats baseline, or a content-pipeline cadence that compounds — /money-skillify writes it as a SKILL.md under the project's local skills directory so future sessions can re-invoke it without re-discovery. Use when the user says: 'this worked, save it', 'codify this', 'turn this into a skill', '把这个固化下来', '存成 skill'.
Automatically activates when over-engineering patterns are detected: (1) Modifying code or files the user did not explicitly ask to change (2) Creating new abstraction layers (class, interface, factory, wrapper) without being asked (3) Adding comments, documentation, JSDoc, or type annotations without being asked (4) Introducing new dependencies without being asked (5) Rewriting entire files instead of making minimal edits (6) Diff scope significantly exceeding the user's request (7) User signals like "too much", "don't change that", "only change X", "keep it simple", "stop" (8) Adding error handling, validation, or defensive code for scenarios that cannot occur (9) Generating tests, configuration scaffolding, or documentation without being asked
World-class MQL5 architect skill for MetaTrader 5. Covers zero-error EA/indicator/script/library development, institutional-grade patterns (CRT, SMC, ICT), prop-firm safety module, ATR risk engine, dashboard panels, CSV logger, multi-timeframe confluence, and backtesting optimization. Compatible with Claude, ChatGPT, Gemini, and other AI assistants.
Synthesize outputs from multiple AI models into a comprehensive, verified assessment. Use when: (1) User pastes feedback/analysis from multiple LLMs (Claude, GPT, Gemini, etc.) about code or a project, (2) User wants to consolidate model outputs into a single reliable document, (3) User needs conflicting model claims resolved against actual source code. This skill verifies model claims against the codebase, resolves contradictions with evidence, and produces a more reliable assessment than any single model.
When the user wants to optimize multiple conflicting objectives, find Pareto-optimal solutions, or balance trade-offs between cost, service, quality, and sustainability. Also use when the user mentions "multi-objective," "Pareto optimization," "NSGA-II," "trade-off analysis," "scalarization," "weighted objectives," "goal programming," or "multiple criteria optimization." For single objective, see optimization-modeling.
Based on the song audio, genre, lyrics, characters, and visual references, complete professional MV director concepting, audiovisual structure, and storyboard design before generation. Use for MV planning, director's statement, song segmentation, narrative/performance/concept/atmosphere structure selection, sound-image relationships, character performance and shot staging, montage, grid storyboards, and upstream design of downstream video prompts. Do not use for prompt optimization of models such as H3/Seedance, image or video generation, API calls, RunningHub submissions, EDL, asset editing, or rendering.
Generate and edit images using Google Gemini image models. Use this skill when the user asks to create, generate, make, or edit images with AI. Supports text-to-image, image editing, style transfer, and multi-image composition. Trigger on requests like "create an image", "generate a picture", "make me a logo", "edit this photo", "add X to this image".
Generate videos using Google Veo models via the nano-banana CLI. Use this skill when the user asks to create, generate, animate, or produce videos with AI. Supports text-to-video, image-to-video animation, dialogue with lip-sync, and scene extensions. Trigger on requests like "create a video", "animate this image", "make a video clip", "generate footage", "produce a short film", "add motion to this".
Analyses a Non-Disclosure Agreement clause by clause and flags unusual terms, one-sided provisions, and negotiation points. Use when reviewing an NDA, mutual NDA, confidentiality agreement, or non-disclosure deed before signing or countering. Produces a plain English verdict, clause-by-clause risk analysis, and a prioritised negotiation checklist — always with a disclaimer that qualified legal advice is required before signing.
Comprehensive news aggregator that fetches, filters, and deeply analyzes real-time content from 8 major sources: Hacker News, GitHub Trending, Product Hunt, 36Kr, Tencent News, WallStreetCN, V2EX, and Weibo. Best for 'daily scans', 'tech news briefings', 'finance updates', and 'deep interpretations' of hot topics.
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.