Spatial World Model (world-model-mcp)

SkillAI & models

Teaches the agent to use the Spatial World Model MCP server to track entities, 3D/2D positions, spatial relationships, object permanence, and movement simulation.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Spatial World Model (world-model-mcp) skill

What this skill tells your AI

The instructions your AI receives, as published by putervision/world-model-mcp in .agents/skills/world-model-mcp/SKILL.md and read by ahel’s review.

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

  1. Inspect World: Call get_spatial_map with format: 'summary' to check total entities and spatial bounding box.
  2. Proximity Lookup: Call query_entities with near_position and max_distance, or entity_id for direct lookup.
  3. Ingest Perception: When new visual elements are detected, call ingest_observation.
  4. Collision Pre-Check: Call simulate_movement before issuing action commands.
  5. Update State: Call record_outcome after actions complete.

3. Tool Reference (15 Tools)

Tool NameKey InputsDescription
update_entityname, type, position, bounding_box, propertiesUpsert entity into spatial memory
query_entitiesquery, type, near_position, entity_id, include_historyFind entities by text search, proximity, or entity ID
set_relationsource_id, relation, target_id, offsetRecord spatial relationship (on, inside, near, etc.)
get_spatial_mapregion_id, format (json, gltf, obj, summary)Export structured spatial layout or summary
simulate_movemententity_id, delta_position, mode, check_collisionsPredict movement path or compute navigation waypoints
ingest_observationobserver_pose, detections, visual_state_id, reconcileIngest perception detections and reconcile frustum view
get_expected_viewobserver_position, observer_orientation, fovCalculate visible entities from observer pose
link_to_goaltask_id, entity_id, relationship, actionAssociate entity with task or extract spatial slice
record_outcomeaction_name, success, resulting_positionUpdate world state after action completion
manage_spatial_specaction, name, bounds, constraintsSpatial SDD contract registration and verification
create_evidence_packtask_id, entity_ids, snapshot_idsImmutable cryptographic evidence package
use_spatial_blackboardaction, topic, sender, payloadMulti-agent coordination and mutex locks
manage_snapshotaction, name, snapshot_a, snapshot_b, entity_idSnapshots, diffs, undo mutations, and time-travel
generate_game_inputsentity_id, target_position, control_profile, actionPlaywright game inputs and screen coordinate projection
wait_for_spatial_stateentity_id, condition, threshold, timeout_msPolling and waiting for spatial state condition

Signals

GitHub stars
16
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
world-model-mcp
Source
github.com/putervision/world-model-mcp