Considerations and Limitations in Multi-Agent Solutions
Review these considerations and limitations for Multi-Agent Orchestration for Agentforce.
Required Editions
| Available in: Lightning Experience |
| Available in: Enterprise, Performance, Unlimited, and Developer Editions. Required add-on licenses vary by agent type. |
- You can connect agents as subagents within a single Salesforce org. Multi-agent
orchestration isn't currently supported for Agentforce agents:
- across multiple Salesforce orgs
- with external, non-Agentforce agents
- You can use these agent type combinations for orchestrator agents and connected
subagents:
- Agentforce Service Agent (ASA) orchestrator agent with ASA connected subagents
- Agentforce Employee Agent (AEA) orchestrator agent with AEA connected subagents
- AEA orchestrator agent with ASA connected subagents
- File-based agents (for example, SDR or Analytics) aren't supported in this release.
- Multi-agent Orchestration users are required to:
- Activate new agent versions to be available as connected subagents.
- Update scripts to use handoff mode. See "Orchestrator Agent Routing to Connected Subagents" in Agent Script in Multi-Agent Solutions.
- Connect only active agents to your orchestrator agent. Don’t select an inactive agent.
- Both orchestrator agents and connected subagents can escalate to a human representative. See "Escalate to a Human Representative" in Agent Script in Multi-Agent Solutions.
- This release supports one level of delegation from an orchestrator agent to connected subagents.
- Timeout limits:
- For orchestrator agents: 120 seconds. The timeout includes the elapsed time spent by the orchestrator agent doing specialized work, such as delegating to connected subagents and synthesizing the results.
- For the connected subagents: 30 seconds.
- Agentforce Voice isn’t a supported configuration for this release of Multi-Agent Orchestration for Agentforce. However, you're welcome to experiment with the understanding that it's unsupported.
- For this release, we support both context variables and custom variables. We support variable mapping in one direction only: from an orchestrator agent to connected subagents.
- Currently, you can't deactivate a connected subagent.
- For customers who didn't participate in the Multi-Agent Orchestration for Agentforce
(Pilot) program, before using an agent with Multi-Agent Orchestration for Agentforce:
- If the agent was built with the new Agent Builder, create and activate a new version of it.
- If the agent was built with the legacy Agentforce Builder, upgrade it to the new Agent Builder. See Upgrade an Agent from the Legacy Builder to the New Builder.
Did this article solve your issue?
Let us know so we can improve!
