Skip to main content
Glama

You:     Redesign ~/Desktop/living-room.jpg in Japandi style, give me 2 options.
Claude:  ⟶ luw_interior_design  ✓ 2 designs in 21s   [shows both images]
You:     Love the first one. Make the sofa green velvet, then turn it into a fly-through video.
Claude:  ⟶ luw_edit_image ⟶ luw_generate_video   ✓

Quick start

Add https://mcp.luw.ai/mcp to your client and sign in to Luw.ai. Nothing to install, no key to copy: signing in creates one for you. New accounts get free credits. To disconnect, delete that key on the API page.

Settings → Connectors → Add custom connector, paste https://mcp.luw.ai/mcp, then Connect and sign in to Luw.ai.

claude mcp add --transport http luw https://mcp.luw.ai/mcp --scope user

Then type /mcp, pick luw and sign in.

Open Luw.ai in ChatGPT, add it and sign in to Luw.ai.

The Luw Studio tools aren't in ChatGPT's directory yet; add them yourself: Settings → Apps & Connectors → Advanced settings, turn on Developer mode, then Create: URL https://mcp.luw.ai/studio, authentication OAuth, and sign in.

Click Add to Cursor above, or add to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "luw": { "url": "https://mcp.luw.ai/mcp" }
  }
}

Cursor shows Needs login next to luw; click it to sign in.

Click Install in VS Code above, or add to .vscode/mcp.json:

{
  "servers": {
    "luw": { "type": "http", "url": "https://mcp.luw.ai/mcp" }
  }
}

VS Code asks you to sign in the first time a tool runs.

URL

https://mcp.luw.ai/mcp

Transport

Streamable HTTP

Auth

Sign in (MCP OAuth), or header Authorization: Bearer YOUR_LUW_API_KEY

For a client that can't sign in, create a key at app.luw.ai/dashboard/api and send it as a header:

{
  "mcpServers": {
    "luw": {
      "url": "https://mcp.luw.ai/mcp",
      "headers": { "Authorization": "Bearer YOUR_LUW_API_KEY" }
    }
  }
}

If it can neither sign in nor send headers, use https://mcp.luw.ai/mcp?api_key=YOUR_LUW_API_KEY. Anyone who has that URL can spend your credits, so keep it private.

The hosted server is stateless and stores nothing; your key goes straight to the Luw.ai API on each request. It can't read files from your computer, so give it image URLs, or use the local setup below.

Local: work with files on your computer

The local server runs on your machine with npx, so it can read paths like ~/Desktop/room.jpg. It needs Node.js 20+ and an API key from app.luw.ai/dashboard/api.

claude mcp add luw --scope user --env LUW_API_KEY=YOUR_LUW_API_KEY -- npx -y @luw-ai/mcp

Or install it as a plugin. Claude Code asks for your key and keeps it in secure storage:

claude plugin marketplace add Luvi-io/luw-mcp
claude plugin install luw@luw-ai

One click: download luw.mcpb, double-click it, paste your API key. No Node.js needed.

Or add this to claude_desktop_config.json (Settings → Developer → Edit Config):

{
  "mcpServers": {
    "luw": {
      "command": "npx",
      "args": ["-y", "@luw-ai/mcp"],
      "env": { "LUW_API_KEY": "YOUR_LUW_API_KEY" }
    }
  }
}

Add to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "luw": {
      "command": "npx",
      "args": ["-y", "@luw-ai/mcp"],
      "env": { "LUW_API_KEY": "YOUR_LUW_API_KEY" }
    }
  }
}

Add to .vscode/mcp.json. VS Code asks for your key and keeps it in its secret storage:

{
  "inputs": [{ "type": "promptString", "id": "luw_api_key", "description": "Luw.ai API key", "password": true }],
  "servers": {
    "luw": {
      "command": "npx",
      "args": ["-y", "@luw-ai/mcp"],
      "env": { "LUW_API_KEY": "${input:luw_api_key}" }
    }
  }
}

Add to ~/.config/devin/mcp_config.json (Windows: %APPDATA%\devin\mcp_config.json). Older Windsurf versions use ~/.codeium/windsurf/mcp_config.json.

{
  "mcpServers": {
    "luw": {
      "command": "npx",
      "args": ["-y", "@luw-ai/mcp"],
      "env": { "LUW_API_KEY": "YOUR_LUW_API_KEY" }
    }
  }
}
codex mcp add luw --env LUW_API_KEY=YOUR_LUW_API_KEY -- npx -y @luw-ai/mcp

Or in ~/.codex/config.toml:

[mcp_servers.luw]
command = "npx"
args = ["-y", "@luw-ai/mcp"]
env = { LUW_API_KEY = "YOUR_LUW_API_KEY" }
gemini extensions install https://github.com/Luvi-io/luw-mcp

Gemini CLI asks for your key and keeps it in the system keychain. Or add to ~/.gemini/settings.json:

{
  "mcpServers": {
    "luw": {
      "command": "npx",
      "args": ["-y", "@luw-ai/mcp"],
      "env": { "LUW_API_KEY": "YOUR_LUW_API_KEY" }
    }
  }
}

Click Add to Kiro above, confirm, then replace YOUR_LUW_API_KEY in Kiro's MCP config.

Windows: if npx isn't found, use "command": "cmd", "args": ["/c", "npx", "-y", "@luw-ai/mcp"].

Related MCP server: no-api-media-mcp

What you can ask

  • "Redesign this bedroom (~/Downloads/bedroom.jpg) as Mid-Century Modern, 3 variations."

  • "This listing photo is an empty room. Stage it as a cozy Scandinavian living room."

  • "Render my sketch plan-sketch.png as a photoreal modern villa at golden hour."

  • "Change the floor in this kitchen to herringbone oak." (segments the floor, then swaps the material)

  • "Remove all the furniture from this room."

  • "Put this chair photo on a terrazzo floor in a sunlit gallery for an ad."

  • "Turn this render into a drone fly-through video."

  • "Make a 3D model of this armchair from these 3 photos."

  • "Seamless blue zellige tile texture, 1024px."

  • "Ask ArchiGPT how to improve the feng shui of this floor plan."

Built-in prompts show up as slash commands or prompt templates in your client: redesign_room, stage_empty_room, sketch_to_render, product_shot.

Tools

Tool

What it does

Credits

luw_interior_design

Redesign a room in any style; empty_room for virtual staging

1

luw_exterior_design

Redesign a facade or building exterior

1

luw_sketch_to_render

Sketch or drawing to photoreal render

1

luw_render

3D, CAD or clay view to photoreal render

1

luw_edit_image

Edit any image with a sentence (Magic Prompt); 2K/4K

1 (2 at 4K)

luw_magic_wand

Masked edit: replace, remove, or apply a material

1

luw_landscape_design

Gardens and outdoor areas, matched to climate and sun

1

luw_upscale_image

Enhance and enlarge 2/3/4× (8× gives 4×)

1

luw_expand_image

Outpaint a cropped photo to a wider view

1

luw_remove_furniture

Empty a furnished room

1

luw_vectorize_image

Photo or drawing to SVG

1

luw_background

Remove a background, or replace it with a generated scene

1

luw_segment

Object masks (all objects, or by prompt) as image URLs

1

luw_generate_image

Text to image (Fluw), or format: "svg" for vectors

2

luw_generate_pattern

Seamless, tileable textures and patterns

1

luw_generate_video

Image to cinematic video with 12 camera motions, optionally ending on an end_image

10 (20 with Symphony)

luw_image_to_3d

Photos to a textured 3D model (GLB)

3 (8 with Symphony)

luw_archigpt

Ask ArchiGPT, an AI architect, a standalone question (accepts images)

1 per ~3k words

luw_get_result

Collect a long-running job (free)

0

luw_list_options

Valid styles, room and building types, camera motions, materials (free)

0

luw_upload_file

Upload a file to Luw.ai storage and get a URL (free)

0

luw_run_model

Call any Luw.ai model with raw API parameters (local server only)

varies

luw_personas

Personas: reusable style identity, training images and slots (local server only)

0

luw_projects

Projects (boards), folders and media (local server only; hosted: luw://projects resources)

0

luw_team

Enterprise: credits, usage, members, invitations (opt-in)

0

Every generation uses your Luw.ai credits. Variations are billed one generation each.

Luw Studio: edit your 3D projects

Luw Studio designs buildings in 3D. With the studio tools, your assistant edits your Studio projects directly: rooms, walls, doors and windows, stairs, roofs, furniture and finishes. Each edit is saved to the project as it is made, and a Studio tab that has the project open shows it within seconds.

  • Hosted: add https://mcp.luw.ai/studio instead of /mcp (same sign-in). It has every hosted tool plus the studio tools.

  • Local: add studio to LUW_TOOLSETS, e.g. LUW_TOOLSETS=generate,archigpt,personas,projects,studio.

"In my Luw Studio project Beach House, add a 3×4 m bedroom east of the living room with a window, and furnish it."

Tool

What it does

luw_studio_list_projects

Your Studio projects, newest first, with each project_id and Studio link

luw_studio_create_project

A new project: an empty site, or a starter house (list_templates)

luw_studio_rename_project

Rename a project

luw_studio_undo

Put back the scene from before the last assistant edit (again: redo)

Scene tools

Luw Studio's editor tools, each taking a project_id: read (get_level_summary, get_zones, get_walls, find_by_type, measure, verify_scene, …) and build (create_story_shell, create_room, add_wall, add_door, add_window, create_stair, create_roof, furnish_room, place_items, paint, apply_patch, …)

The scene tools run on the same core as Luw Studio, with its catalog (Luw's own pieces included) and its defaults. They don't use credits. A project last saved by the previous Luw Studio is readable; open it once in Studio to edit it here. luw://studio/guide is the building guide the assistant reads first.

Resources

Read-only context you can attach to a conversation. In Claude Code, type @luw: to pick one; other clients that support resources offer them as attachments.

Resource

What it holds

luw://account

Your credit balance

luw://history/{app}

Your latest results in one Luw.ai app (Interior, Exterior, Video, …): the 10 newest made through the API and MCP, then your 10 most recently updated projects there with their latest results: file links, prompts and source images

luw://projects/{id}

One project: its folders and the file link of every item in it

luw://catalog/{kind}

Design styles, room and building types, video camera motions, materials and persona professions, with previews

luw://guide

Every tool and what it costs

How it works

  • Local files just work. Any image argument accepts an https:// URL, a local path (~/Desktop/room.jpg), or a data: URI. Local files are uploaded to Luw.ai storage as temporary files (not added to your library) and cached for the session, so repeated edits don't upload the same file again.

  • You see the results. Finished images are embedded in the tool result, so Claude and other clients show them inline. Set LUW_OUTPUT_DIR to also save every result to disk.

  • Long jobs don't time out. A call waits up to 50 seconds and streams progress to clients that show it. If a job like a video takes longer, the tool returns a processing_url and the assistant collects it with luw_get_result, without generating (or paying) twice.

  • Never billed twice. Every generation request carries an Idempotency-Key, so if a network error forces a retry, Luw.ai returns the original job instead of charging again.

  • Masks are handled for you. luw_segment returns masks as URLs, so "change the floor" becomes segment ⟶ magic wand with no manual masking.

  • Fast startup. The package is bundled with zero runtime dependencies, so npx downloads a single ~2 MB package and starts right away.

Configuration

Variable

Default

LUW_API_KEY

(required)

Your API key (get one). LUW_API_TOKEN also works.

LUW_OUTPUT_DIR

(off)

Also download results (images, videos, GLB, SVG) into this folder

LUW_TOOLSETS

all except team and studio

Comma-separated list from generate, archigpt, personas, projects, team, studio, or all

LUW_STUDIO_CATALOG_URL

https://studio.luw.ai/api/catalog

Where the studio tools read Luw Studio's item catalog; empty uses the built-in list

LUW_WAIT_TIMEOUT_SECONDS

50

How long a call waits before returning a processing_url. Raise it for clients with long tool timeouts, such as Claude Code.

LUW_INLINE_IMAGES

true

Embed result images in tool output

LUW_API_BASE_URL

https://api.luw.ai/v2

API endpoint override

Self-hosting the remote server

The same package serves the hosted Streamable HTTP endpoint:

npx -y @luw-ai/mcp --http --port 8080 --host 0.0.0.0   # MCP at /mcp, health at /health
docker build -t luw-mcp . && docker run -p 8080:8080 luw-mcp

It's stateless. Each request brings its own key in Authorization: Bearer …, X-Luw-Api-Key, or ?api_key=. Set LUW_MCP_OAUTH_SECRET to also let clients sign in to Luw.ai instead (OAuth); the sign-in hands the client an ordinary API key. It deploys to Vercel as-is (vercel.json is included), and a Procfile is included for Heroku. See docs/maintainers.md for deployment and release steps.

Development

npm ci
npm test               # unit and protocol tests (no network)
npm run build          # bundles dist/cli.js
npm run inspector      # try every tool in the MCP Inspector
LUW_API_KEY=... npm run smoke            # live checks against the real API (free)
LUW_API_KEY=... npm run smoke -- --paid  # plus one real generation (1 credit)
npm run mcpb           # build/luw.mcpb, the Claude Desktop extension
node scripts/vendor-editor.mjs ../luwstudio2  # rebuild vendor/: the editor core + MCP as Luw Studio runs them

Support

© Luvi Technologies, Inc., all rights reserved. Free to install and run with your Luw.ai account; copying, changing or redistributing it is not allowed (see LICENSE). Bundled open-source components and their licenses: dist/THIRD_PARTY_NOTICES.md.

Available Tools

24 tools
luw_archigptAsk ArchiGPT (architecture & design expert)AInspect

Ask Luw.ai's ArchiGPT, an AI architect: design advice, feng shui, space planning, material and color suggestions, technical questions, and estimating square meters from a photo or plan. Attach an image to discuss it. Each call is a standalone question, so put the details it depends on into the message. Costs 1 credit per ~3000 words.

Inputs: message (required), image, language, persona_id.

luw_backgroundRemove or replace backgroundAInspect

Luw.ai Change Background: cut out a product/object from its background (transparent PNG) — or, with a prompt, place it in a new generated scene for marketing shots ("on a marble kitchen counter, morning light"). Omit prompt to just remove the background. Costs 1 credit.

Inputs: image (required), engine, prompt, enhance_prompt.

luw_edit_imageEdit image with a prompt (Magic Prompt AI)AInspect

Luw.ai Magic Prompt AI: edit any image with a plain-language instruction: "make the sofa green velvet", "add a pendant lamp over the table", "turn it into a night scene". Pass products, materials or a mood board as reference_images to place or apply them. For edits restricted to an exact area, use luw_magic_wand. Costs 1 credit per variation (2 at 4k).

Inputs: image (required), prompt (required), engine, format, +4 more.

luw_expand_imageExpand a cropped photoAInspect

Luw.ai Expand AI: outpaint a tightly cropped architectural or interior photo to a wider view. Costs 1 credit.

Inputs: image (required), format, precision.

luw_exterior_designExterior design (Exterior AI)AInspect

Luw.ai Exterior AI: redesign a building exterior or facade photo — houses, villas, apartments, commercial buildings — in a new architectural style, with new materials, colors and landscaping, keeping the structure. Costs 1 credit per variation.

Inputs: image (required), engine, format, prompt, +8 more.

luw_generate_imageText to image (Fluw AI)AInspect

Luw.ai Fluw AI: generate a photorealistic image or illustration from a text prompt (optionally guided by an input image) — concept art, interiors, products, marketing visuals. Set format="svg" for a vector illustration (Fluw Vector AI). Costs 2 credits per image.

Inputs: prompt (required), seed, image, format, +4 more.

luw_generate_patternSeamless pattern / texture (Pattern AI)AInspect

Luw.ai Pattern AI: generate a seamless, tileable pattern or texture — tiles, wallpaper, fabric, terrazzo, wood, stone — ready to repeat across a surface. Costs 1 credit.

Inputs: prompt (required), size, image, engine, +1 more.

luw_generate_videoImage to cinematic video (Video AI)AInspect

Luw.ai Video AI (Motion): animate a design image into a short cinematic video — fly-throughs, dolly moves, drone shots, reveals. Camera motions: A Forward Dolly, Flythrough Cinematic, Return to Empty Room, Lighting and Color Details, Drone flying to center, Dramatic Zoom Out, Reverse Zoom In, Slow Pull Back, Reveal Zoom, Ground to Sky Tilt, Side Tracking Zoom, Aerial Descent. Pass end_image to make the video end on that frame, e.g. the original photo as image and its redesign as end_image for a before → after transition. Takes a few minutes — expect a processing_url to collect with luw_get_result. Costs 10 credits (aria) or 20 (symphony), with or without end_image.

Inputs: image (required), engine, prompt, end_image, +1 more.

luw_get_resultGet generation resultA
Read-only
Inspect

Collect the result of a Luw.ai generation that was still processing (tools return a processing_url when a job outlasts their wait). Waits for the job to finish (up to the server's wait limit) — call again if it's still running. Free; never re-run a generation instead.

Inputs: processing_url (required), wait, source_image.

luw_image_to_3dImage to 3D model (3DGen AI)AInspect

Luw.ai 3DGen AI: generate a textured 3D model (GLB) from a photo of an object — furniture, decor, products. Add up to 3 photos from other angles for better accuracy. For text-to-3D, first make an image with luw_generate_image. Costs 3 credits (aria) or 8 (symphony, which adds a high-detail texture pack on paid accounts).

Inputs: image (required), engine, texture, simplify, +2 more.

luw_interior_designInterior design (Interior AI)AInspect

Luw.ai Interior AI: redesign an interior photo — home, office, shop, hotel — in any style while keeping the room's architecture. Restyles furniture, materials, colors and lighting. Set empty_room=true to furnish an empty room or turn it into another room type (virtual staging). Costs 1 credit per variation.

Inputs: image (required), engine, format, prompt, +9 more.

luw_landscape_designLandscape & garden design (Landscape AI)AInspect

Luw.ai Landscape AI: design a garden, yard or outdoor area inside a masked region of a photo, with plants chosen for the location's climate and sun exposure. Costs 1 credit.

Inputs: image (required), mask_image (required), city, prompt, +2 more.

luw_list_optionsList styles, room types, materials…A
Read-only
Inspect

Look up valid values for other Luw.ai tools: design_styles (styles param), interior_types (room_type), exterior_types (building_type), video_styles (camera_motion), materials (Magic Wand material_image catalog), professions (persona profession). Free.

Inputs: kind (required), search, details.

luw_magic_wandMasked edit: replace, remove, change material (Magic Wand AI)AInspect

Luw.ai Magic Wand AI: change only a masked area of an image. Give a prompt to add/replace what's there, remove=true to erase it, or material_image to re-surface it (e.g. new flooring or wall tiles). The mask is a black-and-white image the same size as the input; white marks the area to change. Get masks from luw_segment. Costs 1 credit.

Inputs: image (required), mask_image (required), engine, prompt, +4 more.

luw_personasManage personas & their training imagesA
Destructive
Inspect

Personas hold a reusable design identity: style knowledge plus saved images (trainings). Slots 1-6 train the persona's visual style (slot 1 is the style-transfer reference used by style_reference="persona"), slot 7 is the default input image, 8+ are free-form data; add_training without a slot "likes" a design. Pass persona_id to generation tools to use one. Actions: list, get, create, update, delete, list_trainings, add_training, update_training, delete_training, clear_slot.

Inputs: action (required), info, name, page, +10 more.

luw_projectsManage Luw.ai projects, folders & mediaB
Destructive
Inspect

Organize work in Luw.ai projects (called boards in the API), with nested folders and media. Use add_media to save generated results into a project. Actions: list, get, create, rename, delete, list_folders, create_folder, update_folder, delete_folder, add_media, delete_media, move_media.

Inputs: action (required), url, name, extras, +5 more.

luw_remove_furnitureEmpty a roomAInspect

Luw.ai Remove Furniture AI: remove the furniture from a room photo, keeping walls, floor and windows. Costs 1 credit.

Inputs: image (required), format, precision.

luw_renderPhotorealistic render (Render AI)BInspect

Luw.ai Render AI: turn a 3D model view, CAD/BIM screenshot, clay render or basic visualization into a photorealistic architectural render. Costs 1 credit per variation.

Inputs: image (required), format, prompt, precision, +2 more.

luw_run_modelRun any Luw.ai model (advanced)AInspect

Call POST /generate with raw API parameters, for models or options the dedicated tools don't cover (e.g. persona-slot inputs, webhooks). Models: interior, exterior, sketch, render, magicprompt, magicwand, moodboard, landscape, enhance, expand, vector, removefurniture, segment, segmentprompt, fluw, fluwvector, pattern, changebg, removebg, video, 3dgen. Parameter reference: https://luw-ai.gitbook.io/api. Image fields (image, end_image, mask_image, material_image, style_transfer, extra_image_1…6) also accept local paths and data: URIs. Spends credits like the model's own tool.

Inputs: model (required), params.

luw_segmentSegment objects into masks (Segment AI)AInspect

Luw.ai Segment AI: detect objects in a photo and return black-and-white masks as image URLs — every object (wall, floor, sofa, …) or only what you describe in prompt. Feed a mask URL into luw_magic_wand or luw_landscape_design to edit exactly that area. Costs 1 credit.

Inputs: image (required), labels, prompt.

luw_sketch_to_renderSketch to render (Sketch AI)AInspect

Luw.ai Sketch AI: turn a hand sketch, line drawing, floor-plan perspective or rough draft into a photorealistic render. Costs 1 credit per variation.

Inputs: image (required), engine, format, prompt, +7 more.

luw_upload_fileUpload a file to Luw.aiAInspect

Upload an image, video or 3D file to Luw.ai storage and get a URL usable by every Luw.ai tool. Rarely needed: all tools already accept local paths and data: URIs and upload them automatically. Useful to get a shareable URL, to re-host an image from a site Luw.ai can't reach, or to store a file permanently.

Inputs: source (required), persistent, persona_id.

luw_upscale_imageUpscale an imageAInspect

Luw.ai Photo Enhance AI: enhance the quality of an image and enlarge it 2x, 3x or 4x (8x is sent as 4x). Costs 1 credit.

Inputs: image (required), scale, precision.

luw_vectorize_imageConvert an image to SVGAInspect

Luw.ai Vector AI: convert a photo or drawing into a clean SVG vector. Costs 1 credit.

Inputs: image (required).

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.7/5.0

    Scored across 24 tools

    Disambiguation4/5

    Each generation tool maps to a distinct model and input type (exterior vs interior vs sketch vs render vs landscape), and the mask-based tools are separated from the prompt-based ones by explicit description cues. The main soft spots are luw_run_model acting as a raw escape hatch over every other generation tool, and the border between luw_edit_image and luw_magic_wand, which only the descriptions disambiguate.

    Naming Consistency4/5

    A uniform luw_ prefix plus largely verb_noun names (edit_image, upscale_image, remove_furniture, generate_video, get_result) keeps the set readable and predictable. Deviations are noun-only names matching product brands (luw_archigpt, luw_magic_wand, luw_personas, luw_projects, luw_background), and multi-action tools (personas, projects) fold their verbs into parameters instead of names.

    Tool Count3/5

    At 24 tools the surface is heavy, and several entries are largely redundant: luw_vectorize_image duplicates luw_generate_image with format="svg", luw_remove_furniture is a special case of masked editing, and luw_run_model subsumes the whole generation set. Each still maps to a real backend model, so the count is defensible but sits at the top of a comfortable range.

    Completeness5/5

    The lifecycle is well covered: generation (interior/exterior/sketch/render/landscape/pattern/video/3D), editing (edit, magic wand, expand, upscale, background, segment), async collection via luw_get_result, file upload, option lookup, plus personas and projects for reusable identities and saved media. luw_run_model closes any remaining model or parameter gap, leaving only minor omissions like a credits/balance check.

    Maintenance

    ActivityActive
    ResponsivenessLimited data

    Related MCP Connectors

    Related MCP Servers