coding-agent

Run interactive coding agents in PTY-backed background sessions.

1|Updated Apr 12, 2026
One-click install
npx skills add https://github.com/cryptopafi/nexusos-skills --skill coding-agent-cryptopafi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: coding-agent
Source: https://github.com/cryptopafi/nexusos-skills/tree/main/coding-agent
Command: npx skills add https://github.com/cryptopafi/nexusos-skills --skill coding-agent-cryptopafi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the friction of running, monitoring, and coordinating interactive terminal-based coding agents by prescribing PTY-backed background workflows, safe workdir practices, and process monitoring so agents behave reliably and do not interfere with sensitive workspaces.

Core Features & Use Cases

  • PTY enforcement: Ensures interactive CLIs (Codex, Claude Code, Pi, OpenCode) run with pty:true to avoid hangs and broken output.
  • Background orchestration: Launch agents in background sessions, capture sessionIds, and monitor progress with process actions like log, poll, submit, and kill.
  • Safe review and parallel work: Use temporary git repos or git worktrees for PR reviews and parallel issue fixes while avoiding sensitive directories.
  • Notification and lifecycle rules: Add automatic wake triggers on completion, follow strict rules about where not to run agents, and provide clear progress updates.

Quick Start

Start a Codex agent in the background with PTY enabled in your project directory to implement a new feature while I monitor and report progress.

Frequently Asked Questions about coding-agent

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

FAQPage Schema
How do I run interactive CLI coding agents in the background without them hanging?

Run interactive terminal-based coding agents in PTY-backed background sessions to prevent hangs. Enforcing pty:true ensures interactive CLIs receive correct terminal emulation, while background orchestration captures sessionIds for continuous process monitoring.

What is the best way to review pull requests using coding agents without affecting my main workspace?

Review pull requests using coding agents by cloning git repositories or creating temporary git worktrees. This safe workdir practice isolates agent file modifications, ensuring parallel code reviews and issue fixes do not interfere with sensitive directories.

How do I monitor and manage background coding agent sessions after launching them?

Monitor background coding agent sessions using process actions like log, poll, submit, and kill. After launching agents in PTY-backed sessions, capture the sessionId to track progress, submit inputs, or terminate the agent lifecycle safely.

Do I need a specific bash executor to run PTY-backed coding agents?

Yes, you require a bash executor that supports pty:true, background sessions, and process monitoring actions. A git repository context is also required for some agents to perform tasks like PR reviews or large codebase refactoring.

Why do terminal-based coding agents break or stall in standard background executions?

Terminal-based coding agents stall in standard executions because they expect interactive terminal environments. Applying PTY enforcement provides the necessary pseudo-terminal interface, avoiding broken output and unexpected process hangs.

Can I use interactive coding agents to refactor large codebases and build new features?

Yes, you can delegate building new features and refactoring large codebases to interactive coding agents. By running them in PTY-backed background sessions, agents can perform iterative file exploration and execute complex modifications reliably.