skills.rest
erwinkn
erwinknCOMMUNITY
Explore all skills by this author →
Website𝕏@erwinkhnGitHub

ai-agents

Delegate user-specified AI CLI tasks to designated command-line tools.

Updated Jan 14, 2026
Search Tags:#cli#codex#ai#claude#agent#gemini#opencode
One-click install
npx skills add https://github.com/erwinkn/ai-config --skill ai-agents-erwinkn
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-agents
Source: https://github.com/erwinkn/ai-config/tree/main/.agents/skills/ai-agents
Command: npx skills add https://github.com/erwinkn/ai-config --skill ai-agents-erwinkn

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill streamlines delegation to various AI command-line interfaces (CLIs), allowing users to specify a CLI like Codex, Claude, Agent, Opencode, or Gemini for task execution.

Core Features & Use Cases

  • CLI Delegation: Executes tasks using user-specified AI CLIs.
  • Context Management: Supports running CLIs headlessly, in read-only mode, or in isolated worktrees.
  • Use Case: A user wants a specific AI model to review code changes, so the skill provides the command to run codex exec --sandbox read-only 'Review the staged changes in the current repository'.

Quick Start

Use the ai-agents skill to run 'codex exec '<task>'' when a user requests Codex specifically.

Frequently Asked Questions about ai-agents

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

FAQPage Schema
How do I delegate code review tasks to a specific AI CLI like Codex or Claude?▼

You can delegate code review tasks to specific AI CLIs like Codex or Claude by executing commands through the ai-agents skill, which routes your task to the specified command-line interface for processing.

Can I run Gemini or Opencode in read-only mode for code analysis?▼

Yes, you can run Gemini or Opencode in read-only mode for code analysis by passing specific sandbox arguments, ensuring the CLI executes tasks without modifying your repository.

What is the best way to manage context when running AI agents headlessly?▼

The best way to manage context when running AI agents headlessly is to use isolated worktrees or read-only sandboxes, preventing unintended modifications while the CLI processes your delegated tasks.

Do I need a locally installed CLI to use AI agents for task delegation?▼

Yes, you need the specified CLI installed locally or accessible via a remote environment, because the skill delegates tasks directly to the command-line interface for execution.

What are the limitations of delegating tasks to AI command-line interfaces?▼

A limitation of delegating tasks to AI command-line interfaces is that execution depends entirely on the local or remote availability of the specified tool, such as Codex or Gemini.

How does ai-agents work with multiple AI CLIs?▼

The ai-agents skill works with multiple AI CLIs by allowing you to specify which tool, such as Claude, Codex, or Opencode, to use for each individual task, ensuring precise control over execution.

Related Skills

Explore More Software Engineering Skills→
huyhung9630huyhung9630
1

omc-ccg

Synthesizes advice from Codex, Claude, and Gemini for task analysis.

Community
Intermediate
dundasdundas

pr-review-loop

Automate GitHub pull request review, CI checks, code fixes, and merging via CLI and git.

Community
Advanced
mindroom-aimindroom-ai
232

agent-cli-dev

Spawn parallel AI coding agents in isolated git worktrees.

Official
Advanced
tblakex01tblakex01
1

benchmark-models

Benchmark the same prompt across Claude, GPT, and Gemini for latency, tokens, and cost.

Community
Advanced