3D New Era AI

Overview Schema Source Related Servers Discussions Admin

Render a 3D view

render_3d

Read-only

Render a home or floor plan as a 3D PNG with outlines, choosing aerial, elevation, visitor, or section views to inspect walls, rooms, furniture, and parts without a GPU.

Instructions

PNG of the home in 3D (software render with outlines, no GPU needed). view: front|back|left|right|top orthographic elevations — front looks from the plan's bottom edge (large y) toward y=0, back from y=0 toward large y, left from x=0, right from large x; cut=cm makes a section keeping only what is beyond that plane from the viewer (front cut=200 keeps y<200, so the wall at y=0 stays as the backdrop; to remove it look from back), aerial (default; frames the whole building; yaw degrees: 0 from east/+x, 90 from south/plan bottom (default 60); pitch down; zoom >1 farther), visitor (current visitor camera) or cam=i (stored point of view). walls=cutaway drops the walls between the eye and a room to 40 cm, walls=down drops them all; either hides ceilings, roofs and the doors, windows and wall pieces of lowered walls, to see the furniture from the side. piece= draws it alone, seen from its own side (view; aerial is three-quarter), framed the same wherever it stands; part= frames one part of its model; ref=i puts its reference photo beside it, from the photo's side. Keep w/h small.

Input Schema

TableJSON Schema

Name	Required	Description
`h`	No	Height px (default 360, 64..1200)
`w`	No	Width px (default 480, 64..1600)
`cam`	No	Stored point of view index (see cameras)
`cut`	No	Elevations: section plane in plan cm (y for front/back, x for left/right, height for top); what lies between the viewer and it is cut away — walls and pieces alike. front views from large y, back from y=0
`ref`	No	With piece: its reference photo at this index, beside the render, from the photo's side and part
`yaw`	No	Aerial turn, degrees
`part`	No	With piece: frame this part of its model (`model` lists them)
`view`	No	Default `aerial`; `front` to `top` are elevations
`zoom`	No	Aerial distance factor: 1 frames the building, 2 twice as far
`level`	No	Storeys drawn: an id like `lv3` draws it and those below, `all` every one (default: as the editor shows them)
`piece`	No	Only this piece, framed from its own side
`pitch`	No	Aerial height angle, degrees
`walls`	No	`up` (default), `cutaway` or `down`

Output Schema

TableJSON Schema

Name	Required	Description	Default
`mimeType`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed7 schema fields changedv3.0.0Sep 30, 2026

addedInput schema / properties / level

Added value: +{
+  "description": "Storeys drawn: an id like `lv3` draws it and those below, `all` every one (default: as the editor shows them)",
+  "type": "string"
+}

addedInput schema / properties / part

Added value: +{
+  "description": "With piece: frame this part of its model (`model` lists them)",
+  "type": "string"
+}

addedInput schema / properties / piece

Added value: +{
+  "description": "Only this piece, framed from its own side",
+  "type": "string"
+}

addedInput schema / properties / ref

Added value: +{
+  "description": "With piece: its reference photo at this index, beside the render, from the photo's side and part",
+  "type": "integer"
+}

changedInput schema / properties / view / description

Previous value: -"`aerial` (default) or `visitor`"New value: +"Default `aerial`; `front` to `top` are elevations"

addedInput schema / properties / view / enum

Added value: +[
+  "aerial",
+  "visitor",
+  "front",
+  "back",
+  "left",
+  "right",
+  "top"
+]

removedOutput schema / properties / mimeType / description

Removed value: -"Type of the image in the content (image/png)"

First observedv0.1.0Sep 24, 2026

TDQS

A4.5/5.0

Behavior5/5

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

Annotations only declare readOnlyHint and openWorldHint, yet the description discloses the visual consequences of the options: cut removes everything between viewer and plane, walls=cutaway/down hides ceilings, roofs, doors, windows and lowered wall pieces, and 'keep w/h small' hints at resource cost. That is meaningful behavioral context far beyond the annotations.

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

Conciseness3/5

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

It is dense and largely information-bearing, but it is delivered as one unparagraphed block governed by semicolons, mixing output description, view geometry, wall behavior and piece/part/ref interactions. An agent has to parse a wall of text to find the clause relevant to a given parameter.

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

Completeness5/5

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

An output schema exists so return values need no explanation, and the description covers the 13-parameter space including mode interactions (piece + view + part + ref, level, walls vs ceilings). Nothing an agent needs to invoke this correctly appears to be missing.

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

Parameters4/5

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

Schema coverage is 100%, so the baseline is already 3, but the description adds real semantics on top: it defines direction conventions (front views from large y toward y=0, left from x=0), yaw origin and default (0=east, default 60), zoom scaling, and cross-parameter behavior like ref requiring piece. It does not explain defaults for every optional parameter, so it is not exhaustive.

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

Purpose5/5

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

The opening states the concrete deliverable — a PNG of the home rendered in 3D — plus the rendering approach ('software render with outlines, no GPU needed'), which distinguishes it from the sibling render_photo and render_plan tools. The verb, resource and output format are all unambiguous.

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

Usage Guidelines4/5

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

Gives selection guidance across view modes (front/back/left/right/top/aerial/visitor/cam), walls modes and piece/part options, including intent statements like 'to see the furniture from the side' and 'to remove it look from back'. It does not, however, explicitly name and exclude sibling tools (render_photo, render_plan), so it stops short of full when-not guidance.

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