coding-agent

Delegate coding tasks to external coding agents via shell.

636|136|Updated Feb 18, 2026
One-click install
npx skills add https://github.com/swarmclawai/swarmclaw --skill coding-agent-swarmclawai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: coding-agent
Source: https://github.com/swarmclawai/swarmclaw/tree/main/skills/coding-agent
Command: npx skills add https://github.com/swarmclawai/swarmclaw --skill coding-agent-swarmclawai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Delegates coding tasks to external coding agents to accelerate development and code reviews without requiring you to hand-hold in your main workspace.

Core Features & Use Cases

  • Agent-based coding: Interact with Claude Code, Codex, Pi, and OpenCode via shell to perform coding tasks.
  • Scoped workspaces: Run tasks in isolated project directories to avoid workspace contamination.
  • Use cases: Building new features in separate projects, reviewing PRs, refactoring large codebases, and exploratory coding with file exploration.

Quick Start

Ask the coding-agent to perform a coding task in a separate project workspace and report back.

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 external agents via shell?

To delegate coding tasks via shell, you direct commands to external coding agents like Claude Code, Codex, Pi, or OpenCode. This skill automates feature development and PR reviews by running tasks in isolated project directories.

Can I use coding agents to review pull requests and refactor large codebases?

Yes, you can use coding agents to review pull requests and refactor large codebases. The skill delegates these complex tasks to external agents via shell, accelerating development without requiring hand-holding in your main workspace.

How does delegating coding tasks to agents isolate work and prevent workspace contamination?

Delegating coding tasks isolates work by running agent operations in separate project directories. This enforces safe usage and prevents workspace contamination through explicit permission checks and scoped workspaces.

What external coding agents can I interact with through shell commands?

You can interact with Claude Code, Codex, Pi, and OpenCode through shell commands. The skill relies on explicit agent commands and permission checks to prevent unintended operations during exploratory coding or feature development.

Do I need to configure separate directories when running coding agents for new features?

Yes, you need separate directories when running coding agents for new features. The skill isolates work in scoped workspaces to avoid contaminating your main workspace, ensuring safe usage during exploratory coding and code reviews.

When should I not use shell-based agent delegation for coding tasks?

You should avoid shell-based agent delegation when explicit permission checks and isolated directories are unavailable. The skill enforces safe usage to prevent unintended operations, making it unsuitable for uncontrolled or shared main workspace environments.