nested-agents

Orchestrate safe nested sub-agent spawning within a Team pipeline.

8|1|Updated Mar 28, 2026
One-click install
npx skills add https://github.com/bostonaholic/team --skill nested-agents
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: nested-agents
Source: https://github.com/bostonaholic/team/tree/main/skills/nested-agents
Command: npx skills add https://github.com/bostonaholic/team --skill nested-agents

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This skill solves the problem of context bloat and inefficient task delegation by providing strict guardrails for spawning sub-agents within a Team pipeline, ensuring that nested operations remain efficient and non-interactive.

Core Features & Use Cases

  • Context Economy: Optimizes agent performance by offloading bulk reading or verification tasks to specialized sub-agents only when necessary.
  • Safety Guardrails: Enforces strict operational boundaries, such as read-only access for helpers and depth limits, to prevent runaway agent behavior.
  • Use Case: When a researcher needs to verify a complex claim across multiple files, this skill allows them to dispatch a skeptic sub-agent to perform the verification without cluttering the main agent's primary context.

Quick Start

Instruct the current agent to use the nested-agents protocol to dispatch a read-only scout for verifying the identified code findings.

Frequently Asked Questions about nested-agents

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

FAQPage Schema
How do I prevent context bloat when orchestrating sub-agents for large-scale codebase exploration?

You can prevent context bloat by using strict guardrails to spawn nested sub-agents that offload bulk reading or verification tasks, keeping the main agent's primary context clear during complex codebase operations.

What are the safety guardrails for spawning nested agents in a Team pipeline?

Safety guardrails for spawning nested agents include enforcing read-only access for helper agents and applying strict depth limits to prevent runaway behavior during task delegation within the pipeline.

Do I need a specific Claude Code version to enable sub-agent dispatching?

Yes, you need Claude Code version 2.1.172 or higher to enable the Agent tool, which is required for dispatching sub-agents and executing nested orchestration tasks.

How do I dispatch a read-only scout agent for complex code review and security analysis?

To dispatch a read-only scout for complex code review or security analysis, instruct the current agent to use the nested-agents protocol to verify findings across multiple files without modifying the codebase.

When should I use nested sub-agents instead of processing tasks in the main agent context?

You should use nested sub-agents when a task requires verifying complex claims across multiple files, allowing specialized agents to handle the heavy lifting without cluttering the main context.