deep-agents-orchestration

Orchestrate subagents, TodoList planning, and HITL workflows in Deep Agents.

1|Updated Aug 23, 2025
One-click install
npx skills add https://github.com/laceto/financialtools --skill deep-agents-orchestration-laceto
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-agents-orchestration
Source: https://github.com/laceto/financialtools/tree/main/.claude/skills/deep-agents-orchestration
Command: npx skills add https://github.com/laceto/financialtools --skill deep-agents-orchestration-laceto

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Orchestrate subagents, planning, and HITL workflows within Deep Agents to coordinate complex, multi-step tasks with safety checks and auditability.

Core Features & Use Cases

  • SubAgentMiddleware enables delegating specialized tasks to subagents
  • TodoListMiddleware enables structured multi-step planning and tracking
  • HumanInTheLoopMiddleware enforces approvals for sensitive actions
  • Use case: orchestrate a data analysis pipeline that delegates specialized steps and pauses for human review

Quick Start

Configure a deep agent with subagents, TodoList planning, and HITL interrupts, then run a task to observe delegated work, planned tasks, and approval prompts.

Frequently Asked Questions about deep-agents-orchestration

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I coordinate subagents and human-in-the-loop approvals in a multi-step automation workflow?

You can coordinate subagents and human-in-the-loop (HITL) workflows by configuring a deep agent with SubAgentMiddleware for task delegation and HumanInTheLoopMiddleware to enforce approval prompts for sensitive actions.

What is the best way to add structured planning and task tracking to complex agent workflows?

The best way to add structured planning is by using TodoListMiddleware, which enables structured multi-step planning and tracking within Deep Agents to coordinate complex tasks and maintain auditability.

Can I pause a complex automation pipeline to wait for human review before resuming?

Yes, you can pause an automation pipeline for human review by configuring interrupt-resume mechanics and HITL interrupts, which pause delegated work and trigger approval prompts before continuing.

How does thread-id persistence work when orchestrating subagents across multiple steps?

Thread-id persistence maintains state across multiple steps by using interrupt-resume mechanics, ensuring delegated subagent tasks and planned TodoList items remain auditable and synchronized throughout the workflow.

Do I need any external dependencies to set up subagent delegation and safety checks?

No external dependencies are required to set up subagent delegation and safety checks; the orchestration is handled entirely within the Deep Agents framework using built-in middleware components.