Skills
101,174 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Generate article cover images across five dimensions (type, color palette, rendering, text, and mood), combining nine color palettes and six rendering styles. Supports cinematic widescreen (2.35:1), widescreen (16:9), and square (1:1) aspect ratios. Use when the user asks to generate a cover, create an article cover, or make a cover image.
Record Twilio voice calls correctly. Covers the critical distinction between Record verb (voicemail) and Dial record (call recording), dual-channel for QA, mid-call pause for PCI, Conference recording, and the ConversationRelay workaround. Use this skill whenever you need to capture call audio for compliance, QA, or analytics.
Planning skill for marketing and promotional messaging. Use when a developer is figuring out how to set up or architect a marketing campaign on Twilio — channel selection, compliance readiness, audience size, geography, and delivery tracking. Handles open-ended requests like "how do I set up a WhatsApp marketing campaign" or "what's the best way to run promotional SMS." Skip this skill when the developer already knows what they want and is asking for API specs or implementation details.
Advanced TypeScript — type-level programming, library/DSL APIs, declaration merging, modern language features at depth (decorators, using, const T, NoInfer, variance), and feature implementation patterns built on advanced types. Trigger on tasks involving recursive conditional types, infer patterns, mapped-type key remapping, variadic tuples, fluent builders with phantom state, schema-first inference (Zod/Valibot), end-to-end-typed API clients, finite state machines, module augmentation, and library-publishing concerns. Trigger even when the user does not say "advanced" — if the work involves type-level algorithms, library-author API design, or going beyond surface-level uses of TS 5.x features, this is the skill. Assumes the reader has absorbed the `typescript-refactor` skill — this one extends those patterns at depth, never restates them.
This skill should be used when the user asks about “Unity performance”, “optimization”, “GC allocation”, “object pooling”, “caching”, “Update loop optimization”, “memory management”, “profiling”, “framerate”, “garbage collection”, or needs guidance on performance best practices for Unity games
Propagate one validated learning from this session into durable AGENT guidance: extract the smallest useful rule, classify its scope (Global / Claude Code Only / Agent-Specific), store it once in the central agent-learnings file, and sync the regenerated `## Learnings` section into every live agent definition across the CLI trees present. A user-only maintenance workflow that mutates durable agent prompts: it grounds the rule in what this session actually revealed, never invents, confirms scope and wording first, and keeps “Claude Code Only” learnings out of subagent prompts. Use to record an agent-behavior rule so future agents inherit it.
Design or critique an analytics or data dashboard — which metrics earn a place, how to organize and prioritize them, and which chart fits each one. Trigger when designing a dashboard, building an analytics or reporting screen, choosing chart types, or reviewing a dashboard for clarity and information overload.
Use when the user explicitly wants implementation planning before coding, asks to create or revise an implementation plan, supplies requirements with explicit approval evidence, a specification, acceptance criteria, or task list, or has inputs concrete enough to plan but not execute. Do not use for rough unapproved requirements drafting.
Create video ads to exact platform specifications for TikTok, Instagram, YouTube, Facebook, and LinkedIn. Covers dimensions, duration limits, the AIDA framework, and caption requirements. Use for: video ads, social media ads, paid media creative, video marketing, ad production. Triggers: video ad, social media ad, TikTok ad, Instagram ad, YouTube ad, Facebook ad, LinkedIn ad, video creative, ad specs, paid media, video marketing, ad production, Reels ad, Stories ad, pre-roll, bumper ad
Turn ONE topic into a finished Vox-style paper-collage explainer / ad video, end to end with Aliyun Bailian CLI + local ffmpeg — script, collage keyframes, motion, voice-over, music, captions, all automated. Use this whenever the user wants a "Vox style" video, a paper/torn-paper collage animation, a "motion collage", a narrated explainer or short ad built from AI-generated collage posters, a scrapbook-style tribute, or wants to turn a topic / product / person into a punchy narrated collage video — even if they don't say the word "Vox". Also use when reproducing Stav Zilber / rom1trs / Higgsfield-style collage ad workflows. Three input modalities: a topic (B-roll), a talking-head video (A-roll mode), or a single photo of a person/product anchored into the collage (C-roll mode). Triggers: "vox video", "collage video", "motion collage", "paper collage explainer", "make a collage ad", "turn this topic into a collage video", "turn my photo/this product shot into a collage video".
WeWrite topic selection module: crawls trending topics across the web + high-frequency vertical needs + historical performance analysis + search demand, and generates 10 score-ranked topics for 公众号 (including high-frequency angles and evergreen topics). Trigger keywords: 公众号选题、找几个选题、今天写什么、热点选题、热搜选题、爆款选题、选题建议. Requires 公众号/WeChat context; should not be triggered by topic selection needs for 抖音/short video/websites.
Upload one or more videos to YouTube. Use when the user wants to "上传到 YouTube", "发 YouTube", "批量上传", "upload to YouTube", "post videos to YouTube", or publish a finished `final/` directory of MP4s. Reads per-video metadata (title / description / tags) from a sibling `UPLOAD_META.md` file when present (the user's standard Markdown format), or from command-line flags. Works behind a SOCKS/HTTP proxy by using `requests` directly for resumable uploads (the stock `google-api-python-client` MediaFileUpload stalls under this user's proxy setup).
YouTube thumbnail design with specific dimensions, contrast rules, and mobile preview optimization. Covers safe zones, text placement, face expression psychology, and A/B testing. Use for: YouTube thumbnails, video cover images, click-through optimization. Triggers: youtube thumbnail, thumbnail design, video thumbnail, click through rate, ctr optimization, youtube cover, video cover image, thumbnail maker, thumbnail tips, youtube design, video preview image
Use when publishing or updating a Zeabur template to the marketplace. Use when user says "publish template", "update template online", "push template to Zeabur", or "create template on Zeabur". Do NOT use for deploying templates (use zeabur-template-deploy instead). Do NOT use for editing template YAML (use zeabur-template instead).
Expert full-cycle enterprise sales strategist for B2B SaaS. Use when planning sales strategy, pipeline management, deal progression, account planning, competitive displacement, or territory optimization. Covers multi-threading, executive engagement, champion development, buying committee navigation, and complex deal orchestration. Use for enterprise selling, account expansion, land-and-expand, and quota attainment.
When the user wants to deploy AI sales development reps, automate sales qualification, build signal-to-action routing, or design AI agent architecture for sales. Also use when the user mentions “AI SDR,” “AI sales agent,” “automated qualification,” “signal routing,” “sales automation,” “11x,” “Artisan,” “AiSDR,” “AI BDR,” or “autonomous sales.” This skill covers AI SDR deployment, qualification automation, and agent architecture for sales development.
Diagnose Day-2 AKS GPU and KAITO incidents using profile-aware, read-only evidence. WHEN: 'Insufficient nvidia.com/gpu', GPU pod Pending, model-load OOM, DCGM/VRAM, KAITO Workspace not ready, or GPU autoscaling. DO NOT USE FOR: setup (airunway-aks-setup), non-GPU incidents (aks-troubleshooting), standalone VM quota (azure-quotas), or generic cost (cost-analysis or cost-optimization from the optional azure-cost plugin).
Lookup documented AKS fixes only when the prompt includes an exact catalog signature and all of its qualifiers: VMCannotFitEphemeralOSDisk; NodePoolMcVersionIncompatible; 'NodeImageVersion is not accepted'; AKS SkuNotAvailable with size, location, and zone; ZonalAllocationFailed with insufficient zone capacity; OverconstrainedAllocationRequest with listed constraints; nested AKS vmssCSE/CSE VMExtensionError_OutboundConnFail, VMExtensionError_K8SAPIServerConnFail, or VMExtensionError_K8SAPIServerDNSLookupFail; or AllocationFailed with the full cataloged internal-error or insufficient-regional-capacity message. Never use for quota errors, code-only or bare wrappers, generic symptoms, incomplete signatures, or failures outside AKS; use aks-troubleshooting or azure-diagnostics.
Use akshare to retrieve real-time and historical data for Chinese financial markets. Use this skill to look up real-time quotes, historical data, and financial statements for financial products such as A-shares, Hong Kong stocks, U.S. stocks, indices, funds, and futures.
Azure service emulators for local development in Docker (2025). PROACTIVELY activate for: (1) Azurite for Storage (Blob, Queue, Table) emulation, (2) Cosmos DB Linux Emulator container, (3) Event Hubs emulator, (4) Service Bus emulator, (5) Azure Functions Core Tools image, (6) Azure SQL Edge container, (7) Azure App Configuration emulator, (8) connection-string conventions for emulators, (9) seeding emulators with fixtures, (10) running integration tests against emulators in CI. Provides: emulator selection matrix, docker-compose templates, well-known dev connection strings, seed-data patterns, and CI integration recipes.
Builds an Azure portal deep link to a specific blade/menu item for supported Microsoft.ContainerService, Microsoft.Kubernetes, and Microsoft.Compute resources using the aks-search-direct-mid semantic search API. WHEN: 'find where in the Azure portal to do something', 'search the portal for an operation', 'construct a portal link from a search query and a resource'.
Provider-independent production workflow for AI agents creating brand launch films, product reveal videos, manifesto films, campaign hero films, website hero videos, investor/customer launch assets, and social cutdowns. Use when planning, scripting, generating, editing, reviewing, or delivering launch films that must align strategy, claims, brand voice, visual direction, product proof, accessibility, compliance, and multi-platform delivery.
Search code through the most precise semantic backend—prefer tilth MCP, with AST search (`sg`) for structural patterns. Use when the user asks to find a symbol, definition, caller, import, or text pattern, with phrases like “where is X defined”, “what calls Y”, “find all usages of Z”, “trace this function”, “find the TODO comments”, or “search for this error string”. Use even when the user says “grep”, “rg”, “ripgrep”, “ag”, “ack”, “fd”, or “find”. Do not use for reading whole files (use cheez-read), editing code (use cheez-write), or running tests/builds.
Audit and improve CLAUDE.md files in repositories. Use when the user asks to check, audit, update, improve, or fix CLAUDE.md files. Scan for all CLAUDE.md files, evaluate quality against templates, output a quality report, then make targeted updates. Also use when the user mentions “CLAUDE.md maintenance” or “project memory optimization”.
A-share data: quotes, financials, shareholders, fund flows, candlestick charts, concept sectors, top trader lists, and market movements. Use when users query Shanghai, Shenzhen, or Beijing A-shares (for example: 603186 Huazheng New Material, 300476 Victory Giant Technology, 000001 Ping An Bank, top trader lists, or northbound capital flows).
Check and collect LP fees from PancakeSwap V3 and Infinity (v4) positions. Use when the user says “collect my fees”, “claim LP fees”, “how much fees have I earned”, “pending fees”, “uncollected fees”, “/collect-fees”, “harvest LP fees”, or asks about fees from a specific token pair position.
Create one or more intentional git commits from the working tree — grounded in the REAL diff, never a guessed message over blind staging. Reads `git diff HEAD` and branch/log context first, scans the changed files for blockers (secrets, conflict markers, stray or debug artifacts), runs the narrowest relevant checks for the changed surface, stages files EXPLICITLY by name, drafts a Conventional Commit subject, then verifies the resulting hash and any leftover working-tree state. Never `git add -A`/`.`, `--no-verify`, `--amend`, or co-author trailers unless asked; stops on secrets or conflict markers and asks before an ambiguous split rather than reporting a clean tree that isn't. Use only when the user explicitly asks to commit.
Expert competitive intelligence and strategy guidance for B2B/SaaS companies. Use when conducting competitive research, creating battlecards, analyzing win/loss data, building market landscape maps, positioning against alternatives, handling competitive objections, writing comparison pages, or developing sales enablement for competitive situations. Use for competitive monitoring, feature comparisons, and market intelligence gathering.
Compose screen and interaction traps: navigation, adaptive layout, window insets, pagers, toolbars, sheets, list drag and selection, sliders, text fields, settings UI, one-shot effects. Use when layout, gestures or navigation misbehave.
Creates rigorous, validated models of entities, relationships, and constraints for database schemas (SQL, NoSQL, graph), knowledge graphs, ontologies, API data models, and taxonomies. Covers relational, document, graph, event/time-series, and dimensional schema patterns with lifecycle modeling, soft deletes, polymorphic associations, and hierarchies. Use when user mentions "schema", "data model", "entities", "relationships", "ontology", "knowledge graph", or when data structures need formalization.
Use when persisting key-value preferences or small typed settings on Android or KMP with Jetpack DataStore — Preferences vs Typed (Proto/JSON) vs Room selection, the IOException and corruption-recovery error traps, serializers with corruption handlers, and the KMP factory with per-platform file paths. Triggers on DataStore, Preferences, PreferenceDataStoreFactory, DataStoreFactory, preferencesDataStore, Serializer, SharedPreferences-to-DataStore migration, or persistent settings work.
Diagnose Power Apps mobile install/startup failures, npm run dev errors, missing package entry points, or QR/device-opening problems even before an app is loaded. Inspect locally first; restore locked dependencies or restart Metro only with explicit approval, never upgrade packages or patch host/MSAL source. For a running native app, monitor sanitized project-local Metro logs and diagnose runtime errors or silent failures such as empty lists and blank screens. Supports startup-only diagnosis, symptoms, port/platform filters, bounded monitoring, and non-mutating --no-fix mode. Verify the original symptom, not merely installation or quiet logs
Use when the user asks for "deep research", a "research report on", a "comprehensive analysis of", to "look into X in depth", to "write a report on X", or any question needing multi-source synthesis beyond a single search — an autonomous multi-step deep research engine implementing an iterative Think → Search → Extract → Synthesize → Stop loop. The LLM drives every decision: what to search, what's relevant, what's missing, and when to stop. Produces a cited, magazine-quality report with inline citations, category-specific formatting, and research stats. For entity vetting/dossiers use entity-research; for news digests use news-monitoring; for source-grounded Q&A use notebooklm-mode.
Use this skill to push frequently-changing Jetpack Compose state reads (scroll position, animation values, drag offsets) out of the Composition phase and down into Layout or Draw using lambda-based modifiers like Modifier.offset { }, Modifier.layout { }, Modifier.graphicsLayer { }, Modifier.drawBehind { }, and Modifier.drawWithCache { }. Covers the three-phase model (Composition, Layout, Draw), why a state read at phase N invalidates phase N and every phase below, the modifier-phase cheat sheet, and lambda providers (() -> T) for hoisting hot values across composables. Use when the developer mentions every-frame work, scroll jank, animation jank, dropped frames, animated alpha or offset, "the whole subtree recomposes on scroll", Modifier.alpha(state.value), Modifier.offset(x.dp), or graphicsLayer.
Complete developer marketing playbook covering DevRel programs, documentation as marketing, API developer experience, community building, and hackathon strategy. For dev-tool founders who need to reach engineers. Follow @WeiYipei on X.
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.