Skip to content
Cranley

Facility audit

Your building in one pass: doors, restrooms, ways out, life-safety equipment seen, occupant load and cleaning cost, ending in a prioritised fix list with rough costs.

What it answers

Section titled “What it answers”
  • Audit our office for accessibility and life safety
  • What should we fix first in this building, and what will it cost?

For the agent, the tool is described as: For a business auditing its current building or suite (one scan): ADA Title III style checks (32 in door clear width, restroom clearances), egress routes and exit widths, occupant load at 150 ft² per person against the doorways out, smoke alarms, sprinkler heads, exit signs and extinguishers seen on the panoramas, and a cleaning contract price from the takeoff. Returns a prioritised fix list priced from the what-if cost book (low reliability), and says where a fix needs a quote.

Who it is for

Section titled “Who it is for”

Facility manager, Office tenant.

Inputs

Section titled “Inputs”
Input Type Required Default Notes
scan string yes Scan id of the building or suite. Filled from the selected building.
door_width_m number (m) 0.813 Clear width a door needs (32 in, ADA 404.2.3). Saved preference access.door_width_m. Range: more than 0.3, at most 2.0.
read_panoramas boolean true Let the vision model read the panoramas (restroom fixtures, life-safety objects); cached after the first run.
cleaning boolean true Price a cleaning contract from the takeoff.

An input that nothing fills (the selected building, a saved preference, a persona’s preset) becomes a question to the person. The agent asks only for those.

Data used

Section titled “Data used”
Layer Id Needed If it is missing
Doors and windows matterport.openings required The verdict is unknown, with the reason.
360° photos matterport.panos optional The tool runs without it.

What it returns

Section titled “What it returns”

A tool result: a verdict (pass, warn, fail or unknown), metrics, checks and findings, each with its source.

  • Panel blocks: VerdictHeader, MetricRow, MiniChart, CheckList, EvidenceStrip, AssumptionChips, StepLog.
  • Speed: slow. Read-only.

How it works

Section titled “How it works”

Facility audit: one bundle for a business’s own building, ending in a prioritised fix list with rough costs.

Reliability

Section titled “Reliability”

Sources this tool gave on the test cafe:

Where it came from Layer Method Reliability Note
Measured from the scan matterport geometry high Sum of room areas on the scan
Measured from a model model geometry medium IBC 2021 Table 1004.5, business areas
Simulated model simulation low What-if cost book: national averages from cost guides, not a quote, no local labour rate
Simulated model simulation low Your rates times measured quantities; rates are assumptions
Measured from the scan matterport geometry medium Opening width in the wall graph; a hinged door’s leaf takes a few centimetres off it
Measured from the scan matterport geometry medium Matterport wall graph, empty rooms; door widths are the opening in the wall

The tool sets the label for each value. A model never does. The tools give a scan’s wall graph the layer matterport whichever device made the scan, so a phone scan’s geometry carries that label too.

Limits

Section titled “Limits”
  • Needs: doors and windows. On a scan without it, the verdict is unknown and says why.
  • It is slow: progress steps reach the workspace while it runs.
  • A result is a planning aid, not an inspection.

On the made-up test cafe

Section titled “On the made-up test cafe”

This section was written by running the tool when the docs were built, on a hand-made RoomPlan export (no real place was scanned), with no sponsor keys and the network switched off.

Inputs used: door_width_m = 0.813 (preset of the persona “Facility manager”).

Warn 2 fixes found, $1,450 priced (low reliability); occupant load 6 at 150 ft² a person. Cleaning about $144 a month.

Value Result Where it came from Reliability
Scanned floor area 905 ft² Measured from the scan high
Occupant load (150 ft² each) 6 people Measured from a model medium
Priced fixes 1450 $ Simulated low
Fixes found 2 Simulated low
Widen doorways well under 32 in 1200 $ Simulated low
Swing-clear hinges on doors just under 32 in 250 $ Simulated low
Cleaning contract per month 144 $ Simulated low
Check Measured Against Status
Ways out for 6 occupants 2 1 pass
Doorway width, Bathroom to Dining Room 0.76 0.813 m fail
Doorway width, Kitchen to Storage 0.81 0.813 m fail
Doorway width, Kitchen to outside 0.86 0.813 m warn
Bathroom: door clear width 0.71 0.815 m fail

Findings: 3 warn, 3 fail, 1 unknown.

What it reported while it ran:

  1. Scan u_testcafe0001: 4 rooms, 905 ft²
  2. Found 5 doorways and passages in the wall graph
  3. Compared each with 0.813 m: 2 fail, 1 close
  4. door_widths: 2 of 5 doorways are narrower than 0.813 m.
  5. Reading 0 bathroom panoramas for the toilet, tub, shower and grab bars
  6. Measured 1 bathrooms
  7. bathroom_clearance: Every bathroom fails at least one check: door clear width.
  8. Found 2 doorways to outside or unscanned space and 3 connections between rooms

Run it

Section titled “Run it”
threshold run facility_audit --scan <scan-id>
curl -X POST http://127.0.0.1:8787/api/tools/facility_audit/run -H 'content-type: application/json' -d '{...inputs}'

Over MCP the tool is named facility_audit. See MCP server and CLI.