spawn-coder

Delegate coding tasks to a coder-agent via the Task tool.

Updated Dec 25, 2025
One-click install
npx skills add https://github.com/Zeliper/z-coder --skill spawn-coder
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spawn-coder
Source: https://github.com/Zeliper/z-coder/tree/main/.claude/skills/spawn-coder
Command: npx skills add https://github.com/Zeliper/z-coder --skill spawn-coder

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill ensures code changes are authored by coder-agent rather than the main agent, maintaining a clean separation of responsibilities and higher-quality output.

Core Features & Use Cases

  • Delegated Coding: Spawn coder-agent for implementing new features or fixes.
  • Context-Aware Work: Provide the task file path, Step, and codebase findings to guide implementation.
  • Result Summary: Return a concise summary of changes and where they were applied.

Quick Start

When you need to implement work for TASK-001 Step 2, spawn coder-agent with the task file path and a description of the work. The coder-agent will respond with a completed set of changes and a summary.

Frequently Asked Questions about spawn-coder

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

FAQPage Schema
How do I delegate code writing tasks to prevent the main agent from writing code directly?

Delegating code writing uses the Task tool to spawn coder-agent with run_in_background set to true, passing the task file path, Step number, Step details, and codebase-search results. This separation ensures the main agent handles orchestration while coder-agent handles implementation, improving code quality and maintaining clean responsibility boundaries.

What information should I provide when spawning coder-agent for a coding task?

Provide the task file path, current Step number, detailed Step description, and codebase-search results. Include external references and context when available. coder-agent uses this structured input to understand scope, access relevant code patterns, and deliver focused changes aligned with your workflow requirements.

Can I use coder-agent for refactoring, bug fixes, and feature additions in the same workflow?

Yes, coder-agent handles refactoring, bug-fixing, feature-adding, and new code creation within a task-driven workflow. Invoke it via the Task tool for each coding work item, specifying subagent_type: coder-agent and run_in_background: true to process changes while your workflow continues.

What happens after coder-agent completes the coding task?

coder-agent returns a COMPLETED or PENDING_INPUT status with a concise summary of changes and their locations. Report this outcome to builder-agent so orchestration logic can route the result, validate completion, or request clarification if input is pending.

Why should I delegate coding tasks instead of handling them directly in the main agent?

Delegating to coder-agent maintains separation of concerns: the main agent focuses on task routing and decision-making while coder-agent specializes in implementation. This division reduces cognitive load on the main agent, improves code consistency, and enables higher-quality output through focused expertise.