ecos-resource-monitoring

Resource monitoring tracks concurrent API rate limits and queue backlogs to prevent capacity breaches. It checks instance limits and active sessions before agent spawning to ensure system stability and reliable team coordination.

Alert-driven response dispatches notifications via AI Maestro and the agent-messaging system when resource thresholds are crossed. This escalates capacity issues to the Chief of Staff to prevent coordination failures caused by resource exhaustion.

Tracking API rate limits and concurrency requires continuous monitoring that logs metrics at regular intervals. This approach tracks active sessions and queue backlogs to prevent threshold breaches during multi-agent operations.

Agent spawning fails during capacity planning when system resources like CPU, memory, or disk exceed configured thresholds. Running proactive resource checks before spawning identifies insufficient capacity and prevents coordination failures.

Yes, resource monitoring supports capacity planning across multiple projects. It enforces CPU, memory, and disk thresholds, tracks active sessions, and generates reports to ensure safe scaling and reliable coordination.

You should not spawn new agents when CPU, memory, or disk thresholds are breached, or when concurrent sessions and API rate limits reach maximum capacity. Always run a system resource check first and adjust spawning accordingly.