Architecture, component, and runbook documents for the Brain server: a phone-home telemetry service for the M8TRX.AI agent fleet. Use this index to navigate the design and operational docs collected for team review.
Original design spec for the brain telemetry server: goals, ingestion contract, data model, and architecture overview.
→Narrowed MVP scope for the M8trx-managed-service-on-Tailscale reality. Includes the end-to-end fleet flow diagram.
→Hook configuration that emits agent telemetry events from Claude Code sessions to the brain.
→Host-side heartbeat agent: liveness, system metrics, and connectivity health for each customer EC2.
→cloud-init script that prepares a fresh EC2 to run the agent stack and register with the brain.
→Design notes for the deployment runbook: what an operator needs to do to bring up a new customer agent.
→Step-by-step runbook for joining a new customer EC2 to the tailnet and registering it with the brain.
→Operator walkthrough of the GUI-driven deployment path: six clicks from zero to a customer agent on the fleet.
→