3D New Era AI

Overview Schema Source Related Servers Discussions Admin

Read the home

get_home

Read-only

Retrieve the current home state from a 3D floor plan project, with options to filter by ids, room, area, or element types and to reduce payload using summary detail or field selection.

Instructions

Home state. detail=summary is cheapest. Ask for less instead of reading everything: ids=[…] resolves ids (group parts included), room=<id|name> and rect=[[x0,y0],[x1,y1]] read one place, kinds=[walls|rooms|dims|labels|furniture|polylines] and fields=[…] trim each row, parts=true opens groups, ndjson=true prints one element per line so a long answer can be read a slice at a time. Every piece carries bounds (plan box with angle applied) and faces (the side it opens toward). Ids share one counter per version (w1, r2, f3…) and are never reused, so a new version may start at any number.

Input Schema

TableJSON Schema

Name	Required	Description
`ids`	No	Only these ids, group parts included, e.g. `["f833","w24"]`
`rect`	No	Only what meets this rectangle, `[[x0,y0],[x1,y1]]` cm
`room`	No	Only what stands inside this room, by id or name
`kinds`	No	Only these kinds: `walls`, `rooms`, `dims`, `labels`, `furniture`, `polylines`
`level`	No	Storey to read: an id like `lv3`, or `all`. Default: the one shown
`parts`	No	List what is inside groups instead of only counting the parts
`detail`	No	`summary` (counts, bounds, room areas) or `full` (default)
`fields`	No	Keep only these fields of each element; `id` is always kept
`ndjson`	No	One element per line (NDJSON) instead of one JSON object, each line tagged with its kind. Long answers stay readable a slice at a time

Output Schema

TableJSON Schema

Name	Required	Description
`rev`	No	Plan revision
`dims`	No	Dimensions: {id, a, b, len, off?}
`name`	No	Project name
`lines`	No	ndjson=true: the head, then one element per line
`north`	No	Degrees from plan up to north, when set
`parts`	No	Note when parts=true found no groups
`rooms`	No	Rooms: {id, name, pts, m2, …}; with detail=summary, rows [id, name, m²]
`walls`	No	Walls: {id, a, b, t?, h?, arc?, type?, sides?\|left?, right?}
`bounds`	No	detail=summary: [[minx,miny],[maxx,maxy]] cm
`counts`	No	detail=summary: how many of each kind
`labels`	No	Labels: {id, text, at, …}
`levels`	No	Storeys: [id, name, elevation, height, selected, index, viewable, reference]
`warnings`	No	What the read could not do as asked
`furniture`	No	Pieces: {id, cat, at, bounds, faces, wdh?, angle?, elev?, name?, …}
`polylines`	No	Free lines: {id, pts, t, color, …}
`background`	No	The scanned plan under the drawing

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Sep 24, 2026

TDQS

A4.1/5.0

Behavior5/5

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

Beyond the readOnlyHint annotation, the description discloses rich behavioral details: detail=summary is cheapest, ids resolve group parts, room/rect read one place, kinds/fields trim rows, parts opens groups, ndjson prints line-delimited output, and every element carries bounds and faces. It also explains id counter semantics across versions. No contradiction with annotations.

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?

The description is dense but efficient; every clause adds practical guidance. It's a single run-on paragraph rather than structured bullets, which slightly hurts scannability, but it front-loads the most important tip ('detail=summary is cheapest') and wastes no words.

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?

With a read-only annotation, an output schema, and 100% schema coverage, the description adds the missing semantic context: id reuse rules, bounds/faces on every piece, and how to request smaller responses. Nothing needed to call the tool correctly is absent.

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 3. The description adds significant meaning on top of the schema: it explains the cost model (summary is cheapest), the effect of combining filters (room/rect read one place), and how parts and ndjson alter the result shape. This goes beyond the field descriptions.

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 opens with 'Home state' and uses the verb 'read' throughout, making it clear this is a read operation for home data. It is distinguishable from siblings like show_plan or open_home by the emphasis on fetching state rather than rendering or loading, but 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?

The description implies usage: call it when you need home state, and the parameter tips ('Ask for less instead of reading everything') provide operational guidance. However, it doesn't state when to prefer this over sibling tools like show_plan or get variants, nor does it give exclusions.

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