skills.rest
lllooollpp
lllooollppCOMMUNITY
Explore all skills by this author →
GitHub

coding-agent

Execute AI coding agents in background shell processes with PTY support.

39|10|Updated Jan 27, 2026
Search Tags:#automation#development#codex#claude#shell#coding agent
One-click install
npx skills add https://github.com/lllooollpp/clawdbot-cn --skill coding-agent-lllooollpp
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: coding-agent
Source: https://github.com/lllooollpp/clawdbot-cn/tree/main/skills/coding-agent
Command: npx skills add https://github.com/lllooollpp/clawdbot-cn --skill coding-agent-lllooollpp

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill enables programmatic control over various AI coding agents (Codex, Claude Code, OpenCode, Pi) by running them as background processes, streamlining complex coding tasks and automation.

Core Features & Use Cases

  • Interactive Terminal Execution: Run coding agents that require a pseudo-terminal (PTY) for correct operation.
  • Background Task Management: Execute long-running coding tasks in the background, with options to monitor, send input, and terminate sessions.
  • Use Case: Automate the refactoring of a specific module in your project by instructing the coding agent to perform the changes in a background process, monitor its progress, and receive a notification upon completion.

Quick Start

Use the coding-agent skill to run codex exec 'Add error handling to the API calls' in the ~/Projects/myproject directory with PTY enabled.

Frequently Asked Questions about coding-agent

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

FAQPage Schema
How do I run AI coding agents like Codex and Claude Code in background shell processes?▼

You can run AI coding agents like Codex and Claude Code as background shell processes by using a pseudo-terminal (PTY) for interactive CLI execution, enabling programmatic control over long-running tasks.

Can I automate code refactoring by sending instructions to a coding agent via shell?▼

Yes, you can automate code refactoring by instructing a coding agent to perform changes in a specified working directory, monitoring its progress through session polling, and receiving completion notifications.

Does running interactive CLI coding agents require a pseudo-terminal?▼

Running interactive CLI coding agents requires a pseudo-terminal (PTY) for correct operation, ensuring proper terminal emulation when executing agents like OpenCode or Pi in background processes.

What actions are available for managing background coding agent sessions?▼

Managing background coding agent sessions supports actions like list, poll, log, write, submit, send-keys, paste, and kill, allowing full control over interactive terminal execution and task termination.

How do I handle long-running code generation tasks without blocking my terminal?▼

You can handle long-running code generation tasks by executing them as background processes with specified timeouts, then using poll and log actions to monitor progress without blocking your terminal.

Can I specify a working directory when executing AI coding agents programmatically?▼

Yes, you can specify a working directory when executing AI coding agents programmatically, ensuring the background shell process operates within the correct project context for code generation and review tasks.

Related Skills

Explore More Software Engineering Skills→
sentientspritesentientsprite

coding-agent

Execute AI coding agents like Codex CLI and Claude Code via bash commands.

Community
Intermediate
dimetrondimetron
122

pi-check-session-logs

Analyze pi-go session logs for tool-call errors and generate a Markdown report.

Community
Intermediate
lehuuhau1231lehuuhau1231

lint-and-validate

Run linting and type checking across multiple programming ecosystems.

Community
Intermediate
victor-brechbillvictor-brechbill
1

coding-agent

Run AI coding agent CLIs for code generation, review, and refactoring.

Official
Advanced