AgentStack
SKILL verified MIT Self-run

Run Home Scene

skill-sumitaich1998-jarvisvr-run-home-scene · by sumitaich1998

>-

No reviews yet
0 installs
2 views
0.0% view→install

Install

$ agentstack add skill-sumitaich1998-jarvisvr-run-home-scene

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

Are you the author of Run Home Scene? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Run Home Scene

Apply a scene = a named bundle of target device states across the room, in one gesture. Then show the resulting smart_home_panel so the user can fine-tune.

Steps

  1. Resolve the scene from the request to a definition. Built-in examples live

in assets/scenes.example.json (load and adapt; users may override).

  1. Expand the scene into concrete device target states (lights on/level,

thermostat temperature_c, blinds level, etc.).

  1. Gate sensitive steps. If the scene changes a lock toward unlocked or

disarms security, confirm via notify first (see secure-home).

  1. Apply each device target and render the panel reflecting the new

states with show_smart_home; confirm with a short agent.speech.

Scene definition (from assets/scenes.example.json)

{ "id": "movie_night", "name": "Movie Night",
  "devices": [
    { "id": "light_1", "type": "light", "state": { "on": true, "level": 15 } },
    { "id": "blind_1", "type": "blind", "state": { "level": 0 } },
    { "id": "thermo_1", "type": "thermostat", "state": { "on": true, "temperature_c": 21 } } ] }

Output

Resulting smart_home_panel:

{ "widget_type": "smart_home_panel",
  "props": { "room": "Living Room",
             "devices": [
               { "id": "light_1", "name": "Ceiling Light", "type": "light", "state": { "on": true, "level": 15 } },
               { "id": "blind_1", "name": "Blinds", "type": "blind", "state": { "level": 0 } },
               { "id": "thermo_1", "name": "Thermostat", "type": "thermostat",
                 "state": { "on": true, "temperature_c": 21 }, "unit": "C" } ] } }

agent.speech: { "text": "Movie night — lights down, blinds closed.", "final": true }

Bundled resources

  • assets/scenes.example.json — starter scenes (movie_night, good_morning,

leaving, bedtime) you can clone and tweak per home.

Edge cases

  • Unknown scene → offer the closest match or list available scenes.
  • Missing devices → apply what exists; mention what the scene expected but

couldn't find.

  • Security in a scene ("leaving" locks the door & arms sensors) → the

increase-security parts are safe to apply; never auto-unlock.

  • Partial failure → report which devices didn't respond; leave the rest set.
  • Custom scene creation → capture the current panel state as a new scene the

user can name and reuse.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.