ZeqGameAI

Physics engine, procgen, netcode, AI, spatial audio.

• Protocol ID — zeq-game-ai
• Category — Gaming
• Endpoint — POST /api/gaming/ai
• Auth — api-key
• Rate limit — 30/min
• Version — 1.287.0
• Precision — ≤0.1% (KO42-enforced)

What it does

NPC behavior trees and pathfinding with HulyaPulse decision cadence. A* on Zeqond grid, GOAP planning, flocking/steering with R(t) randomness injection for organic feel.

Signature

Request

POST /api/gaming/ai

Param	Type	Required	Default	Description
agents	array	✓	—	[{ id, position, goal, behaviorTree }]
navMesh	object		—	Navigation mesh for pathfinding.
planningMode	string		"bt"	'bt' (behavior tree), 'goap', 'utility'.

Response

{ decisions: [{ agentId, action, path, confidence }], zeqond }

Runnable example

curl -sS -X POST \
  -H "Authorization: Bearer $ZEQ_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "agents": [],
  "navMesh": {},
  "planningMode": "bt"
}' \
  "https://api.zeq.dev/api/gaming/ai"

Integrate

1. Domain solver — compose with KO42 + two additional operators from the matching family for pulse-coherent results.
2. Digital twin — pipe sensor data into this protocol every Zeqond to keep the model phase-locked with the system.
3. Alert threshold — flag results whose error_pct exceeds 0.1% as out-of-spec events for the operations layer.

Seeds

• Near — wrap /api/gaming/ai in a language SDK so builders can call it in three lines.
• Medium — publish a reference integration demonstrating ZeqGameAI alongside a real workload, with pulse-aligned metrics.
• Far — propose ZeqGameAI as an open reference standard so other runtimes can implement it verbatim against the Zeq paper.

Papers

• Zeq paper — https://doi.org/10.5281/zenodo.18158152
• Framework paper — https://doi.org/10.5281/zenodo.15825138

Middleware active. Kernel on the 1.287 Hz HulyaPulse. Awaiting next Zeqond.