MCP server for stage-aware, evidence-backed model routing in Codex. It assigns planning, design, development, review, and verification stages to explicit roles while preserving root-agent accountability and live model evidence.
A vendor-neutral MCP server that enables coding agents to delegate tasks, share context, and work as a team through a shared blackboard and task queue.
An experimental MCP server for MNCS-native development and evidence control, enabling Codex to manage project authority, candidate lineage, evidence gaps, and evaluator-mode boundaries.
A shared context and coordination layer for multiple AI agents over MCP, featuring semantic memory, dependency-aware task DAGs, auto-scheduling, role-based access, real-time push, and a live dashboard.