Skip to content

Camera coverage

Shows what indoor security cameras would see, which doors, windows and entry paths they cover, and the blind spots; proposes camera spots if you have none.

  • Where should I put security cameras?
  • Which doors does this camera miss?

For the agent, the tool is described as: Indoor security camera coverage through the scan’s wall graph: a visibility polygon per camera (field of view and range), which exterior doors and ground-floor windows are in view, how much of each door’s walking route to the bedrooms is on camera, and coverage per room. Give cameras, or leave empty to propose up to max_cameras spots. Returns a coverage lens, camera pins, uncovered openings highlighted, entry routes.

Family planning home security.

Input Type Required Default Notes
scan string yes Scan id. Filled from the selected building.
cameras array Camera positions; empty to let the tool propose them.
max_cameras integer 4 How many cameras the proposal may use. Saved preference security.max_cameras. Range: at least 1, at most 12.
fov_deg number (deg) 110.0 Saved preference security.camera_fov_deg. Range: more than 10, at most 360.
range_m number (m) 9.0 Saved preference security.camera_range_m. Range: more than 1, at most 30.

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.

Layer Id Needed If it is missing
Wall outline matterport.walls required The verdict is unknown, with the reason.
Doors and windows matterport.openings required The verdict is unknown, with the reason.

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

  • Draws on the building: a colour per room; doors, walls or rooms coloured by status; pins on what it found; a route with its narrow points.
  • Panel blocks: VerdictHeader, MetricRow, CheckList, FloorPlanMini, Controls, AssumptionChips, StepLog.
  • Speed: fast. Read-only.

Camera coverage: what indoor security cameras see through the wall graph, which entry points and entry paths they cover, and the blind spots.

Sources this tool gave on the test cafe:

Where it came from Layer Method Reliability Note
Measured from the scan matterport geometry medium 2D sight lines through the scan’s walls; doors assumed open, furniture not modelled

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.

  • Needs: wall outline, doors and windows. On a scan without them, the verdict is unknown and says why.
  • A result is a planning aid, not an inspection.

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: max_cameras = 4 (preset of the persona “Family planning home security”); fov_deg = 110 (preset of the persona “Family planning home security”); range_m = 9 (preset of the persona “Family planning home security”).

Pass The proposed 2 cameras see 5 of 5 entry points. Doors assumed open; furniture not modelled.

Value Result Where it came from Reliability
Cameras 2 Measured from the scan medium
Entry points on camera 5 of 5 Measured from the scan medium
Exterior doors not on camera 0 Measured from the scan medium
Ground floor in view 47 % Measured from the scan medium
Dining Room 53 % Measured from the scan medium
Kitchen 46 % Measured from the scan medium
Storage 45 % Measured from the scan medium
Bathroom 0 % Measured from the scan medium
Check Measured Against Status
On camera, door in dining room 1 1 pass
On camera, door in kitchen 1 1 pass
On camera, window in dining room 1 1 pass
On camera, window in dining room 1 1 pass
On camera, window in dining room 1 1 pass

Findings: 1 warn, 2 pass.

What it reported while it ran:

  1. 2 exterior doors and 3 ground-floor windows to watch
  2. Trying 49 corner mounts and aims
  3. 2 cameras: 5 of 5 entry points in view
threshold run camera_coverage --scan <scan-id>
curl -X POST http://127.0.0.1:8787/api/tools/camera_coverage/run -H 'content-type: application/json' -d '{...inputs}'

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