Show map

Skill sumitaich1998/jarvisvr/skills/navigation/show-map

An AI agentic operating system for mixed reality on the Meta Quest 3 — your own J.A.R.V.I.S. Multi-agent orchestration, multimodal perception (sight/hearing/gaze), 42 holographic widgets, 20 LLM providers.

Install

npx -y skills add sumitaich1998/jarvisvr --skill show-map

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Show an interactive 3D map centered on a place, with markers, or a volumetric globe for world-scale views. Use for "show me a map of…", "where is…?", "put Tokyo on a map", or plotting multiple locations. Triggers: map, show me where, on a map, location of, globe, plot these places, where is.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

2.5 KB, as published. Nobody here has run it

Show Map

Render a map_3d (local/regional) or a volumetric_globe (world-scale) centered on the requested place with optional markers.

Steps

Geocode the place(s) to {lat, lon} (provider; offline use known coordinates).
Pick the widget: city/region → map_3d; multiple far-flung places or "the world" → volumetric_globe with markers/arcs.
Choose zoom + style (streets|satellite|terrain|dark) to fit the content.
Render with show_map / show_globe; the map prefers a surface anchor so it sits on a table.
React to select_marker / zoom / pan events.

Output

map_3d (show_map, props per registry.json):

{ "widget_type": "map_3d",
  "transform": { "anchor": "surface", "position": [0.0,0.05,0.0] },
  "props": { "center": { "lat": 35.6762, "lon": 139.6503 }, "zoom": 11, "style": "satellite",
             "pitch_deg": 50,
             "markers": [ { "lat": 35.6586, "lon": 139.7454, "label": "Tokyo Tower", "color": "#FF5252" } ] },
  "interactions": ["tap","grab","drag","resize","slider"] }

World view (show_globe → volumetric_globe):

{ "widget_type": "volumetric_globe",
  "props": { "style": "earth",
             "markers": [ { "lat": 35.68, "lon": 139.69, "label": "Tokyo", "color": "#FF5252" } ],
             "arcs": [ { "from_lat": 35.68, "from_lon": 139.69, "to_lat": 40.71, "to_lon": -74.0, "color": "#4FC3F7" } ],
             "auto_rotate": true } }

Edge cases

Geocode fails → ask for a more specific place; don't guess coordinates.
Many markers → cluster or raise zoom-out; keep labels legible.
"How do I get there?" → that's wayfind, not just a map.
Indoor / room-scale → maps are for geographic places; use measuring / spatial memory for in-room layout.
Surface available? → if no horizontal surface in client.scene, fall back to a world-anchored map in front of the user.