domains / families / agent-tooling / multi-agent-patterns

multi-agent-patterns

authentic

This skill should be used when designing multi-agent systems that need context isolation, supervisor or swarm coordination, explicit handoffs, parallel execution, or a decision on whether multiple agents are justified.

When to use

• designing multi-agent systems
• context isolation
• supervisor or swarm coordination
• explicit handoffs
• parallel execution
• deciding if multiple agents are justified
• handling single-agent context limits
• scaling agent capabilities

When not to use

• deciding task-model fit
• pipeline shape
• project-level cost
• designing hosted sandboxes
• warm pools
• remote sessions
• background runtime infrastructure
• designing tools each agent exposes

Skill page

Source