Fair Usage Policy
API Quotas, LLM Rate Limiting & System Stability Guidelines • Operon Core Systems Inc. • Governed by Operon Core™ Engine
1. Purpose & Scope
To safeguard system stability, ensure predictable token costs, and prevent denial-of-service degradation across multi-agent pipelines, Operon Core enforces dynamic rate limiting and fair usage thresholds across all client agent runtimes.
2. API Quotas & Concurrency Allocations
| Tier / Agent Service | Concurrent Tool Calls | Token Burst Threshold | Monthly Allocation |
|---|---|---|---|
| Nova (Sales Engine) | Up to 25 parallel sessions | 40,000 tokens/min | Scalable per lead volume |
| Echo (Support Swarm) | Up to 50 parallel chats | 80,000 tokens/min | High-frequency burst pool |
| Vox (Voice Telephony) | Up to 10 simultaneous calls | Sub-500ms audio stream | Trunked SIP minute pool |
3. Automated Circuit Breakers & Loop Prevention
To avoid runaway API charges, every agent governed by the Operon Core™ Engine features an automated loop detector. If an agent enters an unexpected recursion or recursive tool execution (>8 iterative hops without terminal output), the circuit breaker suspends the interaction and notifies our human engineers within 60 seconds.
4. Prohibited Behavior & Abuse
Users may not attempt reverse engineering of agent model weights, inject prompt-injection adversarial payloads, or execute load-testing attacks against shared orchestrators without prior authorization.