coding-agent

Name: coding-agent
Availability: InStock
Author: nexu

Delegate coding tasks to AI agents in background sessions.

3.2k|259|Updated Feb 25, 2026

Search Tags:#coding #ai-agents #git-workflow #development-workflow #pty #background-task

Ready to run

npx skills add https://github.com/nexu-io/nexu --skill coding-agent-nexu-io

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: coding-agent
Source: https://github.com/nexu-io/nexu/tree/main/apps/desktop/static/bundled-skills/coding-agent
Command: npx skills add https://github.com/nexu-io/nexu --skill coding-agent-nexu-io

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Delegates coding tasks to Codex, Claude Code, or Pi agents and runs them in background processes to accelerate software development, PR reviews, and large refactors.

Core Features & Use Cases

Background coding tasks across multiple AI agents for feature development, PR reviews, and refactoring.
PTY-aware execution rules per agent: Codex/Pi/OpenCode require PTY; Claude Code uses non-PTY with bypass permissions.
Safe workspace management: dedicated working directories, scratch repos for experiments, and transparent session logging for progress tracking.
Use Case: when you need to implement a new feature, assess PR changes, or refactor a large module without blocking your main workflow.

Quick Start

Spawn a coding agent in a temporary workspace and start a Codex session to implement a feature.

Frequently Asked Questions about coding-agent

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

FAQPage Schema

How do I delegate coding tasks to AI agents and run them in the background?▼

You can delegate coding tasks to AI agents like Codex, Claude Code, or Pi by spawning background sessions that execute feature development, PR reviews, and large refactors without blocking your main workflow.

What is the best way to manage background coding sessions for large refactors?▼

Managing background coding sessions involves using session control commands like list, poll, log, write, submit, and kill to track progress and safely handle large-scale refactoring in dedicated temporary workspaces.

Does Claude Code require PTY for background task execution?▼

No, Claude Code uses non-PTY execution with bypass permissions, whereas other AI coding agents like Codex and Pi require PTY-aware execution rules for running background coding tasks.

How do I track the progress of an AI agent performing a PR review?▼

You can track AI agent progress for PR reviews through transparent session logging and polling commands, ensuring reproducible environments and clear visibility into background task execution.

Can I use temporary workspaces to safely test AI code generation?▼

Yes, you can use dedicated working directories and scratch repos as temporary workspaces to safely run experiments and isolate AI code generation from your main project files.

When should I not use background AI agents for software development?▼

You should avoid using background AI agents for tasks requiring immediate synchronous feedback or highly interactive debugging, as background sessions are designed for focused, long-running coding workflows.