do

Delegate tasks to a subagent and return only the final result.

1|Updated Jan 18, 2026
One-click install
npx skills add https://github.com/anntnzrb/agents --skill do-anntnzrb
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: do
Source: https://github.com/anntnzrb/agents/tree/main/assets/skills/do
Command: npx skills add https://github.com/anntnzrb/agents --skill do-anntnzrb

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Delegates tasks to a subagent to preserve parent context and returns only the result, reducing cognitive load and keeping responses concise.

Core Features & Use Cases

  • Offload tasks to a subagent to preserve context in the main workflow
  • Return only the final result with no reasoning, intermediate steps, or tool outputs
  • Use case: When you want to split work across agents and get a clean, actionable answer quickly

Quick Start

Delegate the user's request to a subagent and report only the final result.

Frequently Asked Questions about do

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

FAQPage Schema
How do I delegate tasks to subagents while preserving parent context?

To delegate tasks to subagents while preserving parent context, offload the work to a subagent and return only the final result without intermediate steps. This reduces cognitive load and keeps the main workflow concise.

What is parallel execution of independent subagent tool calls?

Parallel execution of independent subagent tool calls means multiple subagents run separate tasks simultaneously. The parent agent delegates work, receives only concise outputs, and avoids exposing internal reasoning from any subagent.

Why does my subagent workflow expose intermediate reasoning steps?

Subagent workflows expose intermediate reasoning when delegation is not enforced. This skill solves that by returning only the final result, suppressing internal reasoning, tool outputs, and intermediate steps to maintain concise responses.

Can I split work across multiple agents and get a clean actionable answer?

Yes, you can split work across multiple agents and get a clean actionable answer by delegating independent tasks to subagents. The parent context is preserved, and only the final result is returned without reasoning clutter.

What's the best way to offload work to a subagent without cluttering responses?

The best way to offload work to a subagent without cluttering responses is to enforce a strict delegation pattern that returns only the result. No internal reasoning or tool outputs are exposed, keeping the parent context clean.