coding-agent

Control coding agents via background PTY sessions with lifecycle management.

1|1|Updated Jan 22, 2026
One-click install
npx skills add https://github.com/kelinfoxy/EZ-Homelab --skill coding-agent-kelinfoxy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: coding-agent
Source: https://github.com/kelinfoxy/EZ-Homelab/tree/main/docker-compose/ez-assistant/skills/coding-agent
Command: npx skills add https://github.com/kelinfoxy/EZ-Homelab --skill coding-agent-kelinfoxy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coding teams need reliable, repeatable control over interactive coding agents (Codex, Claude Code, OpenCode, Pi) to run long tasks in the background with preserved context.

Core Features & Use Cases

  • Background sessions: Start and manage PTY-enabled agent processes in the workspace.
  • Session lifecycle: Poll, log, write, submit, and kill sessions for progress tracking.
  • Use Case: Use for long-running AI-assisted coding tasks such as code reviews, refactors, or data processing pipelines.

Quick Start

Launch a coding agent in a target project directory using PTY in background mode and monitor its session.

Frequently Asked Questions about coding-agent

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

FAQPage Schema
How do I run coding agents like Claude Code or Codex in the background?

Background coding agents are controlled via PTY sessions that support interactive workflows across work directories. You start them with explicit flags to safely launch agent processes like Claude Code or Codex in detached mode for long-running tasks.

What is PTY session management for AI coding assistants?

PTY session management enables programmatic control of interactive coding agents via background pseudo-terminal sessions. It handles session lifecycle operations including polling, logging, writing, submitting, and killing agent processes for long-running coding tasks.

Can I monitor and submit input to a long-running AI coding task?

Yes, you can monitor and submit input to long-running AI coding tasks using controlled polling and input submission commands. The Skill allows you to write to active PTY sessions, poll for progress updates, and submit input to background agent processes.

Does background agent session control require specific terminal setups?

Yes, background agent session control requires PTY-enabled terminals to function. The Skill relies on explicit flags and session commands within these terminals to safely start, monitor, and terminate interactive coding agent processes in the workspace.

What coding agents are supported for background session management?

Supported coding agents include Codex, Claude Code, OpenCode, and Pi. The Skill provides programmatic PTY control to manage interactive workflows and session lifecycles for these specific agents across target project directories.

When should I use background PTY sessions instead of foreground agent execution?

Background PTY sessions are ideal for long-running AI-assisted coding tasks such as code reviews, refactors, or data processing pipelines. They allow you to launch interactive agents, preserve context, and poll progress without blocking your foreground terminal.