🚀 New: Subagents in Managed Agents
Managed Agents can now work as a team to handle complex tasks faster, smarter, and at a lower cost! With the introduction of Subagents, Managed Agents can automatically break complex jobs into smaller, focused tasks, de…
Managed Agents can now work as a team to handle complex tasks faster, smarter, and at a lower cost!
With the introduction of Subagents, Managed Agents can automatically break complex jobs into smaller, focused tasks, delegate them to specialized subagents, and execute independent tasks in parallel.
Each subagent handles a specific responsibility, while the main agent coordinates the work, reviews results, and delivers the final response.
✨ What's New?
🤖 Intelligent Task Delegation
Managed Agents can automatically create subagents based on the complexity of a task.
Each subagent focuses on a specific job with access to only the required tools and information.
Independent tasks can run in parallel for faster execution.
Simply describe your requirements in natural language, and the AI Builder handles the setup.
💰 More Cost-Efficient Execution
Subagents work with smaller, task-specific context instead of processing the entire conversation.
Lightweight models handle focused tasks, while the main agent manages planning and reasoning.
Reduced context usage helps lower execution costs for complex workflows.
Subagent usage is included in the parent agent's run billing.
👀 Complete Visibility into Subagent Activity
Use View subagent activity to see how tasks are delegated and executed.
Inspect individual tasks, execution steps, results, and changes made.
Monitor subagent execution through Activity and Agent Logs.
🔒 Controlled & Secure Execution
Subagents receive only the tools and permissions needed for their assigned tasks.
Read-only tasks cannot modify records.
The main agent reviews subagent-generated customer messages before sending.
Existing app connections are securely shared within the agent's execution.
⚡ Works Across Managed Agents
Supported in chat, triggers, scheduled executions, follow-ups, and API runs.
Works with existing knowledge bases, memory, and custom values.
No additional configuration is required to start using subagents.
🛠️ How to Use
Navigate to AI Agents → Agent Studio → Managed.
Create or open a Managed Agent and describe the task you want it to perform.
The AI Builder automatically determines whether subagents are needed and configures them.
You can also ask the builder to "Split this task into subagents" or "Add a subagent to handle calendar operations."
Run your agent and select View subagent activity to monitor execution.
💡 Example Use Case
Prompt:
"Every Monday, identify opportunities that haven't progressed in the last 14 days, check their recent conversations, and create follow-up tasks for the assigned owners."
Your Managed Agent can delegate this work to specialized subagents:
Opportunity Subagent: Identifies inactive opportunities.
Conversation Subagent: Reviews recent customer interactions.
Follow-up Subagent: Creates follow-up tasks for the respective owners.
The main agent coordinates everything and provides a consolidated summary.
🎯 Why This Matters
Handle more complex workflows: Break large jobs into smaller, manageable tasks.
Faster execution: Run independent tasks in parallel instead of sequentially.
Lower AI costs: Reduce unnecessary context and token usage.
Better reliability: Specialized subagents focus on individual responsibilities.
Full transparency: Track delegated tasks, actions, and execution results.
One Managed Agent. Multiple specialized Subagents. More powerful automation, with less effort and lower cost! 🚀
