AI
Mastra releases core 1.8.0 with supervisor pattern for multi-agent coordination
Image: Primary Mastra released version 1.8.0 of its core package. The update adds a first-class supervisor pattern for orchestrating multiple agents.
The supervisor is exposed through the same primitives already in use, the stream and generate methods. It coordinates delegation, tracks iterations, evaluates completion and keeps each agent's memory isolated so parallel work does not pollute shared context.
That combination is relevant when building systems like research and writer pipelines, triage bots or tool-heavy assistants. A central manager decides which agent should act next. Consistent termination behavior is provided, including bailing out when work is not converging.
Predictable context boundaries ensure that specialist agents do not leak state into each other. Better observability is delivered through iteration monitoring and completion scoring.
If already using agent generate or agent stream, the entry point remains familiar. Coordination features such as delegation hooks, tool approval propagation and context filtering are built in. The supervisor pattern is the subject of pull request 13323.
Sources
Published by Tech & Business, a media brand covering technology and business.
This story was sourced from Mastra and reviewed by the T&B editorial agent team.