Skip to main content
Glama

Exterior design (Exterior AI)

luw_exterior_design

Redesign a building exterior or facade photo in a new architectural style with updated materials, colors, and landscaping while keeping the original structure.

Instructions

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.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "image": "<image>"
    },
    "name": "luw_exterior_design"
  }
}

Input Schema

NameRequiredDescriptionDefault
imageYesPhoto of the building (public https:// link or data: URI; the local npx server also accepts a file path).
engineNoLuw.ai model: symphony (Symphony-3, the default: the higher quality) or aria (quicker, for everyday work). If the user picked a model earlier in this conversation, use that one.symphony
formatNoOutput image format.
promptNoWhat you want, in plain language.
stylesNoDesign style names, e.g. ["Scandinavian"]; for Japandi use ["Japanese Design", "Scandinavian"]. Case-sensitive; see luw_list_options.
precisionNoHow much to change: 80+ subtly, 50-79 moderately (default 50), under 50 dramatically. Applies when prompt is empty, with style_reference, or when furnishing an empty room.
persona_idNoPersona whose saved images to use. Its saved styles and room or building type apply unless you send styles or a type.
variationsNoNumber of alternative designs (1-4, default 1); each is billed as a generation.
building_typeNoBuilding/space type, e.g. "Modern House Exterior", "Luxury Villa Exterior", "Outdoor Patio" (see luw_list_options kind="exterior_types").
enhance_promptNoLet Luw.ai write the design brief when prompt is empty and there is no style_reference. A written prompt is used as written.
style_referenceNoStyle-transfer source: an image whose look to copy (public https:// link or data: URI; the local npx server also accepts a file path), or "persona" to use slot 1 of persona_id.
reference_imagesNoUp to 6 reference images (furniture, materials, products, mood boards) to draw from.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changedv0.3.20
    • addedInput schema / properties / engine / default
      Added value: +"symphony"
    • changedInput schema / properties / engine / description
      Previous value: -"Luw.ai model: aria (default) or symphony (Symphony-3)."New value: +"Luw.ai model: symphony (Symphony-3, the default: the higher quality) or aria (quicker, for everyday work). If the user picked a model earlier in this conversation, use that one."
  2. Changed7 schema fields changedv0.3.18
    • changedInput schema / properties / enhance_prompt / description
      Previous value: -"Let Luw.ai's prompt enhancer expand a short prompt."New value: +"Let Luw.ai write the design brief when prompt is empty and there is no style_reference. A written prompt is used as written."
    • changedInput schema / properties / image / description
      Previous value: -"Photo of the building (https:// URL, local file path, or data: URI)."New value: +"Photo of the building (public https:// link or data: URI; the local npx server also accepts a file path)."
    • changedInput schema / properties / persona_id / description
      Previous value: -"Persona whose saved images and style knowledge to use."New value: +"Persona whose saved images to use. Its saved styles and room or building type apply unless you send styles or a type."
    • changedInput schema / properties / precision / description
      Previous value: -"How strictly to keep the input's structure: 90 = precise, 75 = balanced, 40 = creative."New value: +"How much to change: 80+ subtly, 50-79 moderately (default 50), under 50 dramatically. Applies when prompt is empty, with style_reference, or when furnishing an empty room."
    • changedInput schema / properties / reference_images / items / description
      Previous value: -"Reference image (https:// URL, local file path, or data: URI)."New value: +"Reference image (public https:// link or data: URI; the local npx server also accepts a file path)."
    • removedInput schema / properties / seed
      Removed value: -{
      -  "description": "Fix for reproducible results.",
      -  "maximum": 9007199254740991,
      -  "minimum": 0,
      -  "type": "integer"
      -}
    • changedInput schema / properties / style_reference / description
      Previous value: -"Style-transfer source: an image whose look to copy (https:// URL, local file path, or data: URI), or \"persona\" to use slot 1 of persona_id."New value: +"Style-transfer source: an image whose look to copy (public https:// link or data: URI; the local npx server also accepts a file path), or \"persona\" to use slot 1 of persona_id."
  3. Changed2 schema fields changedv0.3.2
    • changedInput schema / properties / persona_id / description
      Previous value: -"Persona whose saved images/knowledge to use (see luw_personas)."New value: +"Persona whose saved images and style knowledge to use."
    • changedInput schema / properties / styles / description
      Previous value: -"Design style names, e.g. [\"Japandi\"]. Case-sensitive; see luw_list_options."New value: +"Design style names, e.g. [\"Scandinavian\"]; for Japandi use [\"Japanese Design\", \"Scandinavian\"]. Case-sensitive; see luw_list_options."
  4. First observedv0.1.0

TDQS

B3.4/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=false, idempotentHint=false, destructiveHint=false and openWorldHint=true, so the mutation/non-determinism profile is carried structurally. The description adds a genuinely useful cost fact ('1 credit per variation'), which matters because the variations parameter bills per generation. It does not say what the response contains or whether results are async and need luw_get_result.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

One front-loaded sentence: verb and resource first, then scope, then the cost clause. The 'Luw.ai Exterior AI:' prefix is light branding padding, but the em-dash list of building types earns its place by widening the match surface.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With 12 parameters, no output schema, and only partial annotation coverage, the description should say what comes back (generated image(s)) and how variations map to results. It covers purpose and cost well but leaves return behavior and the interplay of persona_id/style_reference/prompt unexplained.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so every parameter is already documented in-schema, and the baseline of 3 applies. The only added meaning is the per-variation billing, which clarifies the cost consequences of the variations parameter; styles, precision, persona_id and style_reference are left entirely to the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description uses a specific verb and resource ('redesign a building exterior or facade photo') and scopes it with building types (houses, villas, apartments, commercial) and the constraint 'keeping the structure'. That is enough to tell it apart from luw_interior_design or luw_landscape_design in practice, though it never names an alternative explicitly.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Usage is implied: reach for this when an existing facade photo should be restyled without changing structure. There is no explicit when-not guidance, no pointer to sibling tools for interior or landscape work, and no statement that an input image is mandatory — that only lives in the schema.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.