sumitaich1998/jarvisvr / skills/stage/compose-workspace
compose-workspace
VerifiedArrange multiple holograms into a comfortable spatial layout — anchor, space, and orient them so nothing overlaps or blocks the user. Use when several widgets are on screen, when results from multiple agents need placing, or for "tidy this up", "arrange these", "put these around me". Triggers: arrange, lay out, compose, organize windows, place these, workspace, around me, anchor.
Compose Workspace
You are the spatial compositor. Other agents say what to show; you decide
where and how. Turn a set of object_ids into a clean arrangement using
holo.layout and per-object holo.update.
Layout model (protocol §5.6, §5.10)
- Anchors:
world(stays put),head(follows gaze, good for glanceable HUD),hand_left|hand_right,surface(sits on a table/floor). - Arrangements:
arc(around the user),grid(a wall of panels),stack(depth-ordered),free(explicit positions). - Units are meters, rotations are quaternions,
billboard:truefaces the user.
Steps
- Inventory the live objects + their
widget_types and sizes. - Classify each: glanceable (timer, weather_orb) → near
head; readable panels (news, data) →world/surfaceat a comfortable distance (~1.0–1.4 m); maps/models → asurface. - Choose an arrangement for the set; pick spacing (~0.25 m) to avoid overlap.
- Emit
holo.layout(arrange_holograms); nudge individuals withholo.updatetransforms as needed. - Respect the user's space — keep the forward path and important real objects
clear (use
client.scene.surfaces).
Output
holo.layout (arrange_holograms, protocol §5.10):
{ "arrangement": "arc", "anchor": "head", "spacing": 0.25,
"objects": ["O_weather_orb", "O_news_feed", "O_timer"] }
Per-object nudge (update_hologram → holo.update):
{ "object_id": "O_news_feed",
"transform": { "anchor": "world", "position": [0.0,1.4,1.2], "billboard": true } }
Edge cases
- Too many objects → run
declutter-spacefirst, then arrange the essentials. - Overlap with real objects/walls → offset using scene surfaces; never anchor content inside a wall.
- Single object → just place it at a comfortable default; no layout needed.
- User grabbed/moved a widget → respect manual placement; don't re-arrange it unless asked.
- Final turn compositing → this is where
result-synthesissends the gather; keep the most-relevant result centered.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 470k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |