delegate

Delegate task steps to sub-agents for parallel execution.

17|1|Updated Jun 1, 2026
One-click install
npx skills add https://github.com/jrcjrcc/deepseek-nyamu --skill delegate-jrcjrcc
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: delegate
Source: https://github.com/jrcjrcc/deepseek-nyamu/tree/main/crates/tui/assets/skills/delegate
Command: npx skills add https://github.com/jrcjrcc/deepseek-nyamu --skill delegate-jrcjrcc

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill addresses the challenge of managing multi-step tasks by effectively delegating specific steps to sub-agents, ensuring efficient parallel processing while maintaining architectural oversight.

Core Features & Use Cases

  • Strategic Delegation: Split tasks into parent reasoning and sub-agent execution for optimized workflow.
  • Session Management: Open, evaluate, and close focused sub-agent sessions with clear boundaries.
  • Use Case: Ideal for coding, research, or verification where tasks can be broken down into smaller, manageable pieces that can be executed in parallel.

Quick Start

To initiate a task, use the agent_open command to create a child session, followed by agent_eval to fetch results.

Frequently Asked Questions about delegate

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

FAQPage Schema
How do I manage multi-step coding and research tasks using parallel processing?

To manage multi-step tasks with parallel processing, you break complex workflows into smaller pieces and delegate specific steps to sub-agents. This allows you to execute coding, research, and verification workflows concurrently while maintaining architectural oversight.

What is the best way to delegate tasks to sub-agents for workflow optimization?

The best way to delegate tasks for workflow optimization is to split tasks into parent reasoning and sub-agent execution. You open a child session with agent_open, let the sub-agent execute the specific step, and then fetch results using agent_eval.

How do I open and evaluate sub-agent sessions for task delegation?

You open and evaluate sub-agent sessions by using the agent_open command to create a focused child session, followed by the agent_eval command to fetch and evaluate the results. Clear session boundaries ensure precise management of the delegated pieces.

Can I use sub-agent delegation for parallel verification workflows?

Yes, you can use sub-agent delegation for parallel verification workflows. It is specifically designed for coding, research, and verification contexts where tasks can be broken down into manageable pieces that are executed in parallel across separate sessions.

When should I not use sub-agents for task delegation?

You should not use sub-agents for task delegation when a task cannot be broken down into smaller, manageable pieces or when precise session management and result evaluation are not feasible. It requires tasks that benefit from parallel execution rather than sequential steps.