codex-parallel-dispatch

Route complex tasks to bounded subagents with deterministic write scopes.

Updated Apr 17, 2026
One-click install
npx skills add https://github.com/ccstudentcc/agent-prompts --skill codex-parallel-dispatch
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex-parallel-dispatch
Source: https://github.com/ccstudentcc/agent-prompts/tree/main/.codex/skills/codex-parallel-dispatch
Command: npx skills add https://github.com/ccstudentcc/agent-prompts --skill codex-parallel-dispatch

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill provides a safe, structured way to split complex tasks into bounded subagents, preventing overlapping writes and complex state management by keeping critical decisions on the main thread.

Core Features & Use Cases

  • Safe delegation: map tasks into independent sidecars with clear ownership and non-overlapping boundaries.
  • Role-based orchestration: enforce agent roles, contracts, and explicit return formats for each subtask.
  • Pattern coverage: supports common collaboration models like explore-then-implement, plan-then-execute, and independent review.

Quick Start

Define the objective and non-overlapping axes on the main thread, then dispatch two bounded workers for disjoint modules and, if needed, a reviewer for final verification.

Frequently Asked Questions about codex-parallel-dispatch

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

FAQPage Schema
How do I prevent overlapping edits when running parallel subagents?

Prevent overlapping edits by routing complex tasks to bounded subagents with deterministic write scopes and explicit boundaries. This orchestration enforces role contracts to ensure safe parallel execution and prevents shared-state conflicts across independent sidecars.

How do I decompose tasks for multi-agent orchestration?

Decompose tasks for multi-agent orchestration by defining non-overlapping axes on the main thread, then dispatching bounded workers for disjoint modules. Each subtask receives explicit return formats and clear ownership to ensure reliable execution.

What collaboration patterns does parallel dispatch support for task decomposition?

Parallel dispatch supports common collaboration patterns including explore-then-implement, plan-then-execute, and independent review. These patterns map tasks into independent sidecars with clear ownership and non-overlapping boundaries for safe execution.

When should I not use parallel subagent delegation?

Avoid parallel subagent delegation for tasks that cannot be decomposed into independent sidecars with explicit boundaries and concrete deliverables. It requires non-overlapping axes and deterministic write scopes to prevent shared-state conflicts across workers.