Use the preference model to triage several blockers/decisions when is unreachable. It may execute only work that is already authorized, reversible, local, and low-impact. Read the avatar files in and follow START.md.
Run the preference model in action mode only inside authority the user already granted. Read the avatar files in and follow START.md. The model is not an authority grant.
Run the preference model in decision-support mode. Read the avatar files in and follow START.md. The result is a fallible hypothesis, not a fact about the user or an authority grant.
AGENTS.md instructions for ellmos-ai/ellmos-controlcenter-mcp, a project described as: MCP control plane for local server discovery, profile management, capability bundles, dashboard, and policy audits.
MCP control plane: local server discovery, profiles, capability bundles, and policy audits. Runs locally from the ellmos-controlcenter-mcp npm package.
Server operations MCP: HTTP health checks, log analysis, deploy dry-runs, mail diagnostics. Runs locally from the ellmos-servercommander-mcp npm package.
Use when the user wants to keep an open triage console for a software project - capturing bugs/change-requests as structured tickets, scoring them, and routing them to the right AI provider/sub-agent for an immediate fix or into the project's task management. Cloud-ready: works across multiple machines sharing a…