Skills
96,770 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Add, create or generate a Dapr Workflow that runs on Diagrid Catalyst — the orchestrator, its activities, and the project wiring. Use for a workflow, orchestration, saga, pipeline or long-running multi-step process. Infers the repo's language
Finds out why a data quality check failed or errored, looks at the rows it matched, and recommends the fix. Also covers a materialized view refresh that the data quality gate did not publish. Use when asked why a check is red, what rows failed, why a view shows "Not published", why a view's data is stale after a refresh, or how to get a blocked refresh published. Trigger terms: failed check, failing rows, errored check, data quality failure, not published, blocked materialization, unpublished refresh, check keeps failing.
Plan a large, multi-session, decision-heavy effort as a persistent decision map of tickets. Use only when the user explicitly invokes $decision-map; it maintains .scratch/<effort>/map.md and child decision tickets until fog clears, then hands off to project-spec
Walk a person from an idea to a Claude Managed Agents agent that runs on its own, and stay with them through debugging. Explains each concept at the moment it matters (Agent, Environment, Session, Deployment, vault, memory store, outcome), asks the real decisions as illustrated AskUserQuestion choices, designs the system prompt in Simplified Technical English, finds MCP connectors, sets up credentials without the secret ever reaching the chat, writes the files, applies them with the `ant` CLI, runs a capped test session, and diagnoses failed sessions from their events. Trigger on "crea un agente en Managed Agents", "managed agents wizard", "quiero un agente que corra solo", "agente programado", "agente always-on", "agente que revise mis repos cada lunes", "mi sesión se quedó pegada", "por qué falló mi agente", "debug my managed agent session", "set up a scheduled Claude agent", "deploy an agent on a cron". Not for the Claude Agent SDK (use agent-sdk-wizard), the raw Messages API, or Claude Code itself.
Senior motion designer that turns any product website, any story/topic, or any song into short, shareable motion-graphics videos (15–90 s showreels, teasers, meme/parody shorts, data stories, and beat edits — fast music-driven hype edits with giant words and no narration) — rendered as real MP4s sized for LinkedIn, Instagram, X and Reels/TikTok, with optional sound design and narration. Starts with a short guided intake (language first — Portuguese, English, Spanish or other — then product, story or beat edit (each with its own questions), look, idea, versions, sound, narration, platform, duration, visuals, CTA, narration engine, voice, captions, hook, retention, pacing, QC), extracts the brand straight from the site, generates several versions in parallel, reveals them in Finder and asks which one is best. Use whenever the user wants a motion video, animated reel, promo/teaser clip, social video, animated explainer or "showreel" for a product, a website, a story/topic, or a beat/hype edit on a song.
Clarify a real project's unresolved decisions from inspected project facts and return a bounded handoff artifact. Use only when the user explicitly invokes $project-clarify after readiness or when a project needs formal clarification; it does not create a SPEC, tickets, or automatically start another user-invoked stage
Turn a formal SPEC into tracer-bullet vertical slices with blocking edges, then publish them as single-file-per-ticket issues on the local markdown tracker. Use only when the user explicitly invokes $project-tickets; it verifies the SPEC handoff, drafts a quiz-able breakdown, and hands execution to the frontier via light-implement
Prepare, verify and attest a first-party Skills release, or audit an existing tag and its installation evidence. Use for a new release, fresh-install verification or release-evidence closeout; execute only the requested stages and authorized targets
Signals scout for the parts of a product people barely use. Measures each clickable element and page against the reach of where it lives, gives each barely used one a likely cause, and reports the ones that are reachable and still not used.
Text-to-speech for voiceovers and spoken output. Renders narration for product, feature, and demo videos as per-scene audio files with exact durations, timing offsets, and captions, and speaks text aloud. Supports the built-in macOS/Linux voice, local Kokoro, OpenAI, ElevenLabs, Gemini, Cartesia, and Deepgram. Use when the user wants a voiceover, narration, or audio for a video or screencast, wants something read aloud or announced, wants an audio file (.wav/.mp3/.m4a), wants to compare or choose voices, or wants to set up TTS providers and API keys.
Use PostHog Autoresearch, agent-driven predictive models: ask "who will do X in the next N days?", let an agent train a model, then score people daily into a person property and `autoresearch_prediction` events. Use when a user wants to predict churn, activity, return, feature adoption, conversion or any event or action per person; create or validate a model from a template; start training or score now; read model quality (holdout vs realized AUC, lift, calibration, likelihood segments, coverage); steer training with suggestions; act on predictions with cohorts, feature flags, workflows or SQL; or diagnose a model that fails to train, scores everyone the same, or stops scoring. Trigger on "autoresearch", "predictive model", "propensity score", "churn prediction", "likely to convert", "who will do X", "predicted_p_", "autoresearch_prediction", "realized AUC", "champion model".
Live WhatsApp triage for the owner’s personal number on an always-on Mac: a local daemon reads WhatsApp Desktop’s own database read-only, a rule gate selects which message bursts warrant a headless Claude run, each run judges the conversation and drafts a reply, and a private board on the owner’s tailnet displays everything. It never sends: the owner presses send. Use when the owner wants it set up or an install continued ("set up whatsapp live", "install whatsapp live"), updated ("update whatsapp live"), operated ("is my whatsapp live running", pause, restart, logs), diagnosed ("why didn't this message show on my board", a failed run, a board that will not open), or removed ("uninstall whatsapp live"). NOT the WhatsApp Business or Cloud API, NOT an unofficial WhatsApp library or browser automation, NOT the export-based breakthrough-whatsapp-triage kit, NOT reading WhatsApp to answer a question (breakthrough-whatsapp-read).
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.
Research identity systems and create or maintain project brand guidelines for consistent implementation, with or without Seenry MCP. Use for brand research, codifying an existing identity or authoring usable brand rules; presentation narrative belongs to deck research
Build, review or research interface motion: press feedback, menus, tooltips, dialogs, sheets, toasts, accordions, tabs, page and shared-element transitions, scroll effects, drag and gestures, number and icon changes. Decides whether something should animate at all, then picks the tool, properties, curve, duration or spring, interruption and reduced-motion behavior, with tested modern-CSS recipes and runnable assets. Also studies recorded motion from Seenry references
Audits whether AGENTS.md and/or CLAUDE.md in the current repository still reflect the project's reality (structure, commands, conventions, scope) and reports discrepancies without editing them. Use when the user runs /audit-agent-context, asks to verify or update agent context documentation, or suspects AGENTS.md/CLAUDE.md is outdated
Customize authorized Android TV / Google TV, LG webOS and Samsung Tizen TVs. Review Android cleanup/launcher changes, LG controls and developer apps, or Tizen Jellyfin/Moonlight WGT builds and installation. Explicit LG SlopBro Root uses a separate gated workflow; no flashing or unrestricted system administration
Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.
Implements a PostHog API (headless) survey inside the user's own codebase so it renders with their UI components and captures the exact events PostHog expects, then verifies the first response arrived. Use when asked to implement a survey, build a headless survey, build a custom survey UI, render a survey in my app, or wire up a survey of type `api` in a browser, server, or custom mobile UI. Covers the posthog-js event contract, server-side SDK capture, and post-implementation verification through MCP. To design or create the survey definition, use creating-surveys instead. For a survey that is implemented but not showing or not recording, use debugging-surveys.
Turn an article or viewpoint into an explanatory video (vertical by default, horizontal on request; 1.5–3 minutes, narrated with Ray's replicated voice + subtitles): prioritize real footage (screen recordings, live-action footage, screenshots). Recent episodes mainly use HyperFrames with a style tailored to the content (music, a digital avatar in the lower-left corner, and a publishing package), and can also use a hand-drawn approach rendered frame by frame in code. Use for “turn this into an explainer video,” “make an educational video,” “make a short educational video,” or “narrate it in my voice.” For narration scripts, use ray-kb; for covers, use ray-cover. For Ray explainers and educational videos, prioritize this skill instead of using general hyperframes or faceless-explainer directly (this skill invokes HyperFrames internally).
Whole-codebase deep audit. Partitions the repo into 25-file components, runs 9 perspectives by 3 models (OpenAI, Gemini, Claude) on every component via opencode run with Serena navigation, sweeps for gaps, and runs one ocr scan coverage pass. Then collates and deduplicates the findings and verifies every Critical and High issue against a clean export of the reviewed commit with isolated claude -p judges. The run directory feeds the review-tickets skill.
Turn verified issues from a review-deep, review-full, or review-diff run into Jira ticket files. Re-cuts each issue into one ticket per file from the reviewers' own text, judges and edits every finding against a clean checkout with isolated claude -p processes, merges tickets that share one fix, assembles Jira wiki tickets with code copied from git, and checks them. Ends with posting guidance for any Jira client and a bundle ticket for deferred low-priority work.
【Suite-internal skill · Users usually don’t need to invoke the full workflow directly; just describe the request to the orchestrator short-drama-director, which will route it here automatically】【漫剧老李 V7.4.26 suite · sub-skill】Specialized engine for storyboarding dramatic and everyday scenes (short-drama visual style): converts original novel/script text into minimalist written storyboard scripts using a short-drama visual camera system, shot economy, restrained acting, five dimensions of within-scene continuity, and five camera-movement modes; also produces storyboard images (★ standard image-generation asset): one multi-panel image per segment, divided by seven action phases (wind-up/process/contact/apex/result/reaction/exit), with panel count determined by action complexity rather than a mechanical 3×3 grid; exit frame = next segment’s entry frame; CLEAN_GRID clean version for image generation + optional technically annotated version. ★ Must detect the I art-style lock for the visual foundation (live action/2D/3D/ink wash) and adapt facial close-up angles accordingly—3D/photorealistic styles must not use the “close-up × straight-on eye-level” combination; use angled side views instead (live action observes a 1/3 quota; 2D uses the same setup no more than twice consecutively). Trigger terms: 分镜、文字分镜、电影分镜、生成分镜、帮我分镜、导演分镜、镜头脚本、storyboard、镜头衔接、运镜衔接、故事板、故事版、分镜图、故事板图、面板链、四宫格故事板、四格故事板、六宫格故事板、九宫格故事板、2×2 故事板. This skill is part of the suite; enter the complete workflow through the orchestrator short-drama-director.
Design and build web interfaces at design-studio quality, not AI-template quality. Use for any new page, landing page, dashboard, app shell, pricing, settings, form or single component, web or mobile (native app screens with seenry-apps), for redesigns and visual polish, and for matching a reference. Designs the brand's identity first, the way studios do (a truth, a type voice, one color field, one graphic device, one image treatment), studies real identities, studio sites and category leaders from Seenry MCP, builds the page as an application of that identity on a proven product and motion floor, and does not finish until a strict slop check passes and a blind critic scores the page at big-company level
Design and build native-feeling mobile app screens and flows (SwiftUI, UIKit, React Native, Expo, Flutter, mobile web prototypes) that match top App Store apps. Use for onboarding, tab and stack navigation, lists, forms, settings, paywalls, players, sheets and empty or error states. Researches whole journeys in real apps via Seenry MCP (the flow from the app it is based on, the look from the app it should look like), then enforces navigation doors, a mechanical slop pre-flight and a beat-the-reference gate
Find, generate and integrate the visual material a design needs: photos, hero images, album or product art, PNG and 3D-style icons, UI icons, illustrations, textures and fonts. Sources license-clear assets from the internet (CC0/public domain photos, permissive icon sets, open fonts) and generates images when the host has an image model. Records provenance for every file. Use whenever a page or component would otherwise ship with placeholders, gradients or stock clichés
Review a web interface, a UI diff or a single component and return ranked, fixable findings. Use for design critique, UI audits, accessibility checks, reviewing a branch or PR for interface impact, stress-testing one component across states and sizes, or explaining how an observed interface or animation is built
Direct and render short product films as code: launch videos, feature explainers, tool and agent demos, social cuts. Studies launch videos and recordings in Seenry for pacing, writes the story, shot list and voiceover, builds a seekable HTML film from real captures, renders it frame by frame, designs and mixes the sound, and checks format, loudness and every on-screen claim. Not for editing live-action footage
Use when a session needs events from any polled system delivered as messages instead of polls, or when writing the source that polls it, so one daemon the session owns carries each poll's alerts in the peer envelope, keeps its notes in the log, and quotes nothing from the polled system.
Plan and run a dependency-ordered software change through sequential, one-shot Pi workers with bounded retries and explicit dependency handoffs. Use for authorized multi-step implementation work that benefits from isolated worker contexts; avoid it for planning-only requests or small changes that one worker can complete directly.
Operate Veer-managed worktrees, folder contexts, terminals, repos, automations, artifacts, skill sharing, worktree comments, and Veer's embedded browser through the `veer` CLI. Use when the user says "$veer-cli", "Veer worktree", "child worktree", "spawn codex/claude in a worktree", "read/wait/send Veer terminal", "handoff" / "handover" / "give this to another agent", "Veer browser", "orca artifacts", or "share skills". Prefer it over raw git worktree, ad hoc PTYs, or Computer Use when Veer state is involved. Use Computer Use only when a visible window needs GUI control that a CLI, filesystem, or API cannot do.
Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes: Standards (does the code follow this repo's documented coding standards?) and Spec (does the code match what the originating issue/spec asked for?). Runs both reviews in parallel sub-agents and reports them side by side. Use when the user wants to review a branch, a PR, work-in-progress changes, or asks to "review since X".
Master design skill for Claude Code. Use whenever the user asks to design, redesign, mock up, prototype, build a UI, build a landing page, build a dashboard, build a mobile app screen, generate a design system, generate design tokens, build a component library, clone a website's design, extract a design system from a URL, convert a screenshot to code, push or pull from Figma, critique or review a design, or hand off design to another AI. Triggers on slash commands /design web, /design mobile, /design system, /design landing, /design review, /design import, /design redraw. Also triggers on natural-language phrases like "design this", "redesign", "create UI", "build interface", "clone [url]", "design system", "component library", "create prototype", "hi-fi", "mockup", "canvas", "design tokens", "figma sync", "review this design", "critique", "handoff". Always use this skill for any frontend visual design work — do not fall back to generic frontend-design unless the user explicitly opts out of design-engine.
Comprehensive reference for authoring FlyDSL GPU kernels on AMD GPUs. Covers the layout algebra, tiled copy/MMA, buffer ops, loop-carried range loops, SharedAllocator (LDS), autotuning, and common patterns. Use when writing, reviewing, or understanding FlyDSL kernel code.
Test how coding agents discover and use your product. Use when asked to run agent evaluations, test whether agents can use an SDK, API, CLI, or MCP server, find where agents struggle with onboarding, or measure whether docs and skills improve agent success. Runs real agent sessions and scores outcomes through the Gauge CLI. Also use for Gauge Agents tasks; not ordinary unit tests or AI-search visibility tracking.
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.