Design & 3D modeling

KYC / eKYC MCP server for face compare, liveness, OCR, and image labeling.

AI media built for production. AI orchestration for image, video and audio, in any AI app.

Live weather anywhere, with the radar drawn as text characters, not an image.

Start here for hosting a game on Edgegap (dedicated servers + matchmaking) when the engine is not Unity, Unreal or Godot, is unknown, or the question is about the platform itself. Use when someone wants to host or deploy a game server, wants a single-player or co-op game always online or harder to crack, asks how players will connect to matches, wants servers for a custom-engine game, compares hosting options, or asks what Edgegap apps, versions, deployments, fleets or matchmakers are, even if they never say "hosting". It picks the server shape, explains the vocabulary and safety rules, and routes to the right skill. For a Unity project, load edgegap-unity instead (NGO: edgegap-ngo, FishNet: edgegap-fishnet, PurrNet: edgegap-purrnet, Fusion: edgegap-fusion); Unreal: edgegap-unreal; Godot 4: edgegap-godot; Colyseus: edgegap-colyseus; Cocos: edgegap-cocos; Nakama: edgegap-nakama; Manus: edgegap-manus; Three.js/web: edgegap-threejs; vibe-coded: edgegap-vibe-coded; matchmaking design: edge

Package any game server (Unreal, Godot, Node, Go, Rust, C++, a Unity build) into a Docker image that runs correctly on Edgegap, push it to the Edgegap Container Registry or Docker Hub, GHCR, ECR, GCR or GitLab, and register it as an app version: the container contract, Dockerfile patterns, linux/amd64 builds on Apple Silicon, image tags, registry credentials on the version, and port mapping. Use whenever writing or fixing a Dockerfile for a game server, building on a Mac, pushing to registry.edgegap.com, getting "unable to pull image" or exec format errors, or when someone asks how to get a server build onto Edgegap. Engine-agnostic; edgegap-unity has a ready Unity Dockerfile.

Add in-game voice chat with EOS Voice (Epic Online Services) to an Unreal Engine game that runs on Edgegap dedicated servers. The game server is the trusted server: it checks each player's EOS ID token, puts them in a match or team room from the Edgegap matchmaker data, gets room tokens from the EOS Voice Web API (EOS voice doesn't run on Linux servers) and removes players who leave. Ships tested C++ drop-ins (device ID login so players need no Epic account, a replicated component that needs no game class edits, server token service), Developer Portal client policies, Edgegap env vars, local two-client test with fake audio, and fixes for pitfalls. Use for "add voice chat", "team voice", "proximity or party chat", "EOS voice", "push to talk", "players can't hear each other", Vivox replacement, or voice on an Unreal game hosted on Edgegap. Needs a working Edgegap dedicated server first (edgegap-unreal). Not for Unity yet, lobby-only voice without a server, or Mumble.

Godot 4 + Edgegap: take a Godot multiplayer game (MultiplayerAPI, @rpc, ENetMultiplayerPeer, WebSocketMultiplayerPeer, MultiplayerSpawner, netfox) to dedicated servers and matchmaking, end to end. Load this before using any Edgegap MCP tool on a Godot project (project.godot, .gd, .tscn, export_presets.cfg). It has tested GDScript templates (a server lifecycle autoload that waits for the matched players and stops its own deployment, a matchmaker client and Find Match button), the Edgegap Godot plugin build (export, image, push from the editor), app version ports (UDP for ENet, WS + TLS Upgrade for web and wss), matchmaker config, and fixes for Godot pitfalls: empty server logs, export allowlists that drop new scripts, join validation that silently blocks matched players. Use when a Godot dev wants online play: "host my Godot game", "nobody should host", "Find Match button", "Godot web multiplayer", or is debugging Godot deployments, ports or matchmaking on Edgegap.

Take an Edgegap-hosted game from "the test deployment works" to live players: the launch checklist (token hygiene, app version settings, restart policy, max duration, self-stop, image caching, log storage), version naming and blue/green rollouts with zero matchmaking downtime, cost and leak control, rate limits, free-tier limits to lift before launch, observability (logs, metrics, deployment IDs), capacity planning with Edgegap, Private Fleets and persistent servers, and incident basics. Use when preparing a launch, beta or playtest on Edgegap, reviewing production readiness, rolling out a new server build, deploying builds from CI or GitHub Actions (edgegap/deploy-action), or when someone mentions going live, launch day, rollout, scaling, hosting costs or operating game servers with real players.

Manus + Edgegap: turn a browser game built with Manus (manus.im, Manus Game Dev, a Manus web app, or its code exported to GitHub) into an online multiplayer game with an authoritative dedicated server and matchmaking on Edgegap. Load this before using any Edgegap MCP tool on a Manus project. It has templates: Steam sign-in verified on the backend, backend join/status/cancel/leave routes that create Edgegap matchmaker tickets (mode, region, party, player IP) and return only the server address and a signed join token, a Node.js WebSocket game server that self-stops, a browser client with reconnect, a Dockerfile, a GitHub Actions image build, and a matchmaker config. Use when a Manus user wants online play: "make my Manus game multiplayer", "add matchmaking", "Find Match", "Steam login for my web game", "host my HTML5/JavaScript game server", or is debugging wss, CORS, tickets or deployments for a Manus game. Other Three.js/web games without sign-in: edgegap-threejs. Unity: edgegap-unity.

Netcode for GameObjects (NGO) + Edgegap: take a Unity game built on NGO to dedicated servers and matchmaking, end to end. Load this before using any Edgegap MCP tool on an NGO project (com.unity.netcode.gameobjects, NetworkManager, UnityTransport, StartHost, ConnectionManager, Boss Room). It has NGO C# templates verified on Edgegap, no scene edits (server start and self-stop, Find Match panel, settings asset), a Boss Room adapter, the Edgegap Unity plugin for build, image, push and deploy, port 7777 UDP app version, matchmaker config, and fixes for NGO pitfalls: host-only logic, connection payloads and approval, UnityTransport needing an IP, client-only code on the server, Relay leftovers. Use when an NGO dev wants online play: "host my NGO game", "nobody should host", "replace Unity Relay or Multiplay", "Boss Room on Edgegap", "Find Match button", or debugs NGO deployments, ports or matchmaking. Mirror or WebGL-first games: edgegap-unity. FishNet: edgegap-fishnet. PurrNet: edgegap-pur

Design an authoritative dedicated game server that fits per-match hosting on Edgegap: who owns the game state, the deployment lifecycle (boot, read injected env vars, bind a fixed port on 0.0.0.0, accept players, self-stop at match end, drain on SIGTERM), logging, and what never goes in the server. Use whenever adding online multiplayer to a game, converting a host/client, listen-server or peer-to-peer game to dedicated servers, or when someone asks about netcode architecture, authoritative servers, match lifecycle, tick rate, cheat prevention or how a game server should behave on Edgegap. Engine-agnostic (Unreal, Godot, Node, custom engines); apply it before containerizing. Unity projects get the same rules, already applied, from edgegap-unity.

Diagnose why a game server on Edgegap will not start, keeps restarting, stops early, or cannot be reached, for any engine: image push and pull failures (401, 403, 424 "unable to pull image"), exec format errors, 422 couldn't allocate, Error deployments, crash loops and OOM kills, 100% CPU, deployments stopping after 60 minutes or 24 hours, missing logs, clients timing out or failing to connect (internal vs external port, protocol, 0.0.0.0 binding, wss and TLS Upgrade), and matchmaker players stuck searching. Use when someone says their Edgegap deployment, server or matchmaking "isn't working", or pastes an Edgegap error. Unity-specific build and SDK issues are in edgegap-unity.

Turn raw footage into a finished edit in a specific genre/format — montage, documentary, tutorial, news, workshop, vlog, trailer, explainer. Use whenever the user wants the video "edited into a X", asks "what style should this be / what would make a good X", or hands over raw footage without spelling out every cut. This is the director: it understands the footage first, picks a treatment, then composes the other skills (edl-edit, cutting-rhythm, filler-removal, captions, reframe-social, video-transitions, video-overlay, audio-edit, color-grade, remotion-graphics, grok-video-edit) to execute it.

Turns Claude into a video editor's apprentice. Claude asks the editor to teach it how they work with each client, splits that workflow into stages, works the stages the editor hands over (one at a time, under their corrections), learns from their final cut, and writes everything into an editor skill plus one skill per client. Works with any editing or motion app (Premiere Pro, After Effects, DaVinci Resolve, Final Cut Pro, CapCut and others) through project files, scripting, MCP servers or computer use, in any language. Use it whenever a video editor, montage editor, motion designer or مونتير wants to teach Claude their workflow, onboard a new client or channel, start or continue an episode or project for a client, compare their final cut with Claude's work, or build/update a skill for a client or for themselves — even if they never say "skill" or "apprentice".

Build a multi-clip edit as an auditable JSON "edit decision list" (EDL) — a list of clips with source/in/out and a written rationale per pick — that one command renders to a finished video. Use whenever you're stitching several cuts/takes/segments into one video and want the edit to be reviewable, diffable, and re-runnable instead of a throwaway ffmpeg filtergraph. Covers the schema, the rationale discipline, multicam cutaways, and verifying the output by re-transcribing it.

Interactive ideation and brainstorming skill for the Elephant & Goldfish Model (EGM). Helps turn half-baked concepts into structured hypotheses via clarifying questions, review, and iterative refinement. Produces an RFC summary for human review or more detailed design with egm-draft.

Interactive ideation and brainstorming skill for the Elephant & Goldfish Model (EGM). Helps turn half-baked concepts into structured hypotheses via clarifying questions, review, and iterative refinement. Produces an RFC summary for human review or more detailed design with egm-draft.

Local-first MCP asset QA for headless Blender: bounded scripts, FBX reimports, four-view renders.

Design, review or rewrite a survey or questionnaire so busy people want to answer it AND the answers are worth having; stakeholder and partner surveys, user-needs assessments, post-workshop or kickoff follow-ups, customer and community feedback forms, Google Forms / Typeform / Qualtrics drafts. Use whenever someone shares a survey draft, asks to "make this survey better / shorter / less boring / more engaging", asks what questions to put to users or partners, is turning a meeting or kickoff into a follow-up form, or calls a questionnaire long, dry, or a slog; even if they never say "survey" (e.g. "what should we ask the working groups", "draft the form for the stakeholders"). Not for bias-only wording audits of a finished instrument (a linter does that) or for interview guides alone.

Transforms vague UI ideas into polished, Stitch-optimized prompts. Enhances specificity, adds UI/UX keywords, injects design system context, and structures output for better generation results.

Use for any landing page, hero section or scroll-driven web design that should feel cinematic (parallax, scrollytelling, overlapping sections, text reveals, Apple-style animation), before coding. Inspects assets with inspect-assets.py, then builds with 45+ 2.5D techniques, no WebGL.

Videos, images, voices and songs from eroq's named and uncensored engines, with your characters.

Comprehensive toolkit for protein language models including ESM3 (generative multimodal protein design across sequence, structure, and function) and ESM C (efficient protein embeddings and representations). Use this skill when working with protein sequences, structures, or function prediction; designing novel proteins; generating protein embeddings; performing inverse folding; or conducting protein engineering tasks. Supports both local model usage and cloud-based Forge API for scalable inference.

AI product photos, videos and descriptions for e-commerce, from one product image.

Design & price modular raised garden beds and walls: exact bill of materials, EUR price + checkout.

Short-form video intelligence: your TikTok, Reels and Shorts data, reports, Studio, live editor.

Use when a result could be contaminated because the thing being judged had contact with its own answer key. Fires on: an eval or test suite that keeps passing and feels suspiciously circular, and you want to know if it's testing anything real; being tempted to peek at a held-out or test set "just to calibrate" a threshold or rubric; scoring a model's output using that same model's own prior generations as the answer key; having written the reference, gold, or expected answers yourself and now also judging against them; or assembling training/design data alongside eval data in the same pass with no wall between them. This governs the standing, project-wide separation between what builds a thing and what grades it — not a single judging session. Not for blinding one head-to-head comparison at judging time (use blind-evaluation), and not for defining what counts as admissible in the first place (use validity-floors).

Expert in event sourcing, CQRS, and event-driven architecture patterns. Masters event store design, projection building, saga orchestration, and eventual consistency patterns. Use PROACTIVELY for event-sourced systems, audit trail requirements, or complex domain modeling with temporal queries.

Use when working with *.excalidraw or *.excalidraw.json files, user mentions diagrams/flowcharts, or requests architecture visualization - delegates all Excalidraw operations to subagents to prevent context exhaustion from verbose JSON (single files: 4k-22k tokens, can exceed read limits)

Plans, creates, and optimizes marketing campaigns including content strategy, social media, email, and analytics. Helps develop go-to-market strategies, campaign messaging, and performance measurement.

Explains how a website, a visual effect or an animation was built, from a live URL or a screenshot.

Find out why code or infrastructure is the way it is (a design choice, a threshold, a workaround, or a regression) from git history, GitHub PRs and issues, ADRs, phase docs, CHANGELOG, CLAUDE.md/AGENTS.md, and memory, and report each claim at a stated confidence (found, inferred, unknown) with a citation. Read-only. A Sonnet digger works the git and PR history. Before a change, it ends with a Preserve / Change / Avoid / Risk set. Triggers on "/explain-why", "why does X work this way", "why did we pick Y", "where did this number come from", or "is it safe to remove this". Optional leading level. Use explain-how for runtime behavior.

Review Software Factory failures across GitHub, GitLab, Jira, and Bitbucket bindings; propose and verify repository runner image or sandbox configuration repairs. Use for periodic admin maintenance or repeated needs-human outcomes.

Review Software Factory failures across GitHub, GitLab, Jira, and Bitbucket bindings; propose and verify repository runner image or sandbox configuration repairs. Use for periodic admin maintenance or repeated needs-human outcomes.

Plan, review, build, and deploy a repository-specific Software Factory sandbox image using repository history and verification requirements. Use during factory setup or when onboarding another repository.

Plan, review, build, and deploy a repository-specific Software Factory sandbox image using repository history and verification requirements. Use during factory setup or when onboarding another repository.

Official faktry MCP server: generate and edit images, video, audio, and PDFs.

Unified media generation via fal.ai MCP — image, video, and audio. Covers text-to-image (Nano Banana), text/image-to-video (Seedance, Kling, Veo 3), text-to-speech (CSM-1B), and video-to-audio (ThinkSound). Use when the user wants to generate images, videos, or audio with AI.

Unified media generation via fal.ai MCP — image, video, and audio. Covers text-to-image (Nano Banana), text/image-to-video (Seedance, Kling, Veo 3), text-to-speech (CSM-1B), and video-to-audio (ThinkSound). Use when the user wants to generate images, videos, or audio with AI.

Generate and edit images, video, audio and 3D with fal. Discover models and check prices.

Turn ideas into fully formed designs and specs through natural collaborative dialogue. Use when planning new features, designing architecture, or making significant changes to the codebase.

Command-line audio workstation: stems, restoration, tuning, sheet music, mastering. JSON everywhere.

Local FFmpeg video and audio tools over stdio. Footage stays on this machine.

Convert, probe, and split video into frames — real FFmpeg, paid per job with x402 on Base.

Offline Figma MCP server: layout, SVG assets, renders and revision diffs from saved .fig files

Use the Figma MCP server to fetch design context, screenshots, variables, and assets from Figma, and to translate Figma nodes into production code. Trigger when a task involves Figma URLs, node IDs, design-to-code implementation, or Figma MCP setup and troubleshooting.

Use the Figma MCP server to fetch design context, screenshots, variables, and assets from Figma, and to translate Figma nodes into production code. Trigger when a task involves Figma URLs, node IDs, design-to-code implementation, or Figma MCP setup and troubleshooting.

Connects Figma design components to code components using Code Connect mapping tools. Use when user says "code connect", "connect this component to code", "map this component", "link component to code", "create code connect mapping", or wants to establish mappings between Figma designs and code implementations. For canvas writes via `use_figma`, use `figma-use`.

Generates custom design system rules for the user's codebase. Use when user says "create design system rules", "generate rules for my project", "set up design rules", "customize design system guidelines", or wants to establish project-specific conventions for Figma-to-code workflows. Requires Figma MCP server connection.

Create a new blank Figma file. Use when the user wants to create a new Figma design or FigJam file, or when you need a new file before calling use_figma. Handles plan resolution via whoami if needed. Usage — /figma-create-new-file [editorType] [fileName] (e.g. /figma-create-new-file figjam My Whiteboard)