coding-agent

Delegate coding tasks to AI agents via managed background sessions with workspace isolation.

1|Updated Apr 23, 2026
One-click install
npx skills add https://github.com/karaabd23-crypto/openclaw --skill coding-agent-karaabd23-crypto
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: coding-agent
Source: https://github.com/karaabd23-crypto/openclaw/tree/main/skills/coding-agent
Command: npx skills add https://github.com/karaabd23-crypto/openclaw --skill coding-agent-karaabd23-crypto

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Delegates coding tasks to AI agents via a managed background workflow, freeing developers from long-running, repetitive coding sessions and enabling focused, isolated work.

Core Features & Use Cases

  • Orchestrates tasks across Codex, Claude Code, OpenCode, and Pi agents with PTY-safe execution.
  • Supports starting, monitoring, and controlling background coding sessions (start, poll, log, write, submit, kill) within a dedicated workspace.
  • Facilitates rapid code experiments, reviews, and feature iterations without blocking the main development environment.

Quick Start

Spawn a background Codex session to refactor the authentication module in a Git repository.

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 Claude Code in background sessions?

Manage background coding sessions using start, poll, log, write, submit, and kill commands to control AI agents within an isolated workspace, enabling non-blocking monitoring of long-running code refactors and feature iterations.

Does this coding agent tool support Codex and OpenCode with PTY execution?

Yes, this coding agent tool supports Codex, Pi, and OpenCode by enforcing PTY-safe execution, ensuring interactive terminal environments are maintained correctly during background code generation and review tasks.

What is the best way to run multi-agent collaboration for large refactors across projects?

The best way to run multi-agent collaboration for large refactors is orchestrating Codex, Claude Code, and Pi agents simultaneously across temp-workspaces, using session management commands to coordinate isolated background execution.

Can I use background AI agents for code reviews without blocking my main development environment?

Yes, you can use background AI agents for code reviews without blocking your main development environment, because the orchestration workflow isolates tasks in dedicated workspaces for rapid asynchronous iteration.

How do I monitor and control a background Codex session spawned for feature development?

Monitor and control a background Codex session using poll to check status, log to view output, and kill to terminate, providing full lifecycle management over delegated feature development tasks.