bmad-party-mode

To orchestrate multi-agent discussions, you initiate party mode to load a coordinated agent roster, enabling natural group conversations and cross-talk for brainstorming and decision-making sessions.

Multi-agent cross-talk in group conversations works by dynamically selecting 2-3 relevant agents per topic from loaded manifests, enabling them to interact naturally with each other during brainstorming and evaluation sessions.

You load and merge agent personalities from manifests by initiating the session, which automatically coordinates the agent roster loading process to form a diverse discussion team for your conversation.

Yes, multi-agent orchestration is designed specifically for decision-making sessions requiring coordinated input from diverse BMAD agents, satisfying end-to-end orchestration needs from agent selection to graceful exit handling.

To handle graceful exit and session state, the orchestration process tracks the conversation state throughout the session and supports clean termination, ensuring repeatable multi-agent conversations.

No specific dependencies are required to run multi-agent BMAD orchestration, as the skill operates independently to load agent manifests, manage session state, and facilitate group discussions without external component requirements.