world-model-mcp
Skill24 starsChanged 49 days agoMITView on GitHub
Teaches the agent to use the Spatial World Model MCP server to track entities, 3D/2D positions, spatial relationships, object permanence, and movement simulation.
4 other projects use a version of this.
Adopted by
Most recently taken up by
- putervision/vision-memory-mcpword for word, first seen Oct 8, 2026
- putervision/state-memory-mcpword for word, first seen Sep 28, 2026
- putervision/world-model-mcpword for word, first seen Sep 26, 2026
- putervision/agent-reasoning-mcpword for word, first seen Sep 26, 2026
Other versions4 projects use a version of this.
Word for word · 4
- putervision/state-memory-mcp★ 114100% same
- putervision/world-model-mcp★ 50100% same
- putervision/vision-memory-mcp★ 48100% same
- putervision/agent-reasoning-mcp★ 22100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository15 other files it gives its agents.
- .agents/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- .cursor/rules/agent-reasoning-mcp.mdcCursor rule
- .cursor/rules/behavior-mcp.mdcCursor rule
- .cursor/rules/state-memory-mcp.mdcCursor rule
- .cursor/rules/vision-memory-mcp.mdcCursor rule
- .cursor/rules/world-model-mcp.mdcCursor rule
- docs/llms.txtllms.txt
- agent-reasoning-mcpSkill
- behavior-mcpSkill
- state-memory-mcpSkill
- video-ingestSkill
- vision-memory-mcpSkill
- webcrypt-mcpSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 0 are distinctive, and 5 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 20
Inspect World: Call get_spatial_map with format: 'summary' to check total entities and spatial bounding box.
- Very specific: line 21
Proximity Lookup: Call query_entities with near_position and max_distance, or entity_id for direct lookup.
- Very specific: line 22
Ingest Perception: When new visual elements are detected, call ingest_observation.
- Very specific: line 23
Collision Pre-Check: Call simulate_movement before issuing action commands.
- Very specific: line 24
Update State: Call record_outcome after actions complete.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Spatial World Model (world-model-mcp)
This skill provides step-by-step guidance and operational patterns for interacting with @putervision/world-model-mcp.
1. Role in the PuterVision Triad
- Perception Layer (
vision-memory-mcp): Ingests images, detects bounding boxes, extracts OCR. - World Model Layer (
world-model-mcp): Maintains persistent 3D/2D coordinates, bounding volumes, topological relations, and object permanence. - Action/State Layer (
state-memory-mcp): Manages task execution DAGs, decisions, blockers, and milestones.
2. Core Operational Sequence
- Inspect World: Call
get_spatial_mapwithformat: 'summary'to check total entities and spatial bounding box. - Proximity Lookup: Call
query_entitieswithnear_positionandmax_distance, orentity_idfor direct lookup. - Ingest Perception: When new visual elements are detected, call
ingest_observation. - Collision Pre-Check: Call
simulate_movementbefore issuing action commands. - Update State: Call
record_outcomeafter actions complete.
3. Tool Reference (15 Tools)
| Tool Name | Key Inputs | Description |
update_entity | name, type, position, bounding_box, properties | Upsert entity into spatial memory |
query_entities | query, type, near_position, entity_id, include_history | Find entities by text search, proximity, or entity ID |
set_relation | source_id, relation, target_id, offset | Record spatial relationship (on, inside, near, etc.) |
get_spatial_map | region_id, format (json, gltf, obj, summary) | Export structured spatial layout or summary |
simulate_movement | entity_id, delta_position, mode, check_collisions | Predict movement path or compute navigation waypoints |
ingest_observation | observer_pose, detections, visual_state_id, reconcile | Ingest perception detections and reconcile frustum view |
get_expected_view | observer_position, observer_orientation, fov | Calculate visible entities from observer pose |
link_to_goal | task_id, entity_id, relationship, action | Associate entity with task or extract spatial slice |
record_outcome | action_name, success, resulting_position | Update world state after action completion |
manage_spatial_spec | action, name, bounds, constraints | Spatial SDD contract registration and verification |
create_evidence_pack | task_id, entity_ids, snapshot_ids | Immutable cryptographic evidence package |
use_spatial_blackboard | action, topic, sender, payload | Multi-agent coordination and mutex locks |
manage_snapshot | action, name, snapshot_a, snapshot_b, entity_id | Snapshots, diffs, undo mutations, and time-travel |
generate_game_inputs | entity_id, target_position, control_profile, action | Playwright game inputs and screen coordinate projection |
wait_for_spatial_state | entity_id, condition, threshold, timeout_ms | Polling and waiting for spatial state condition |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.