Agent skill

coding-agents

Guide for invoking external coding agents (Codex, Claude Code, Gemini) from the CLI to review code, perform tasks, or get second opinions. Use when the user wants to delegate work to another AI agent, run code reviews externally, or needs help choosing between available coding agents.

View SKILL.md on GitHub Repository
Stars 33
Forks 3

Install this agent skill to your Project

npx add-skill https://github.com/vaayne/agent-kit/tree/main/skills/coding-agents

Metadata

Additional technical details for this skill

os 
[
    "darwin",
    "linux"
]

SKILL.md

Coding Agents CLI Guide

Agent Overview

Agent Command Default Model Best For
Codex codex gpt-5.4 Code review, sandboxed exec
Claude Code claude Opus 4.6 Multi-file tasks, agentic workflows
Gemini gemini Gemini 2.5 One-shot prompts (not installed)

cc is a local shell alias for claude with Bedrock + bypass permissions. Use claude when cc is unavailable.

Claude Code Models (claude --model)

Alias Use When
opus Complex refactors, architecture, multi-file
sonnet Default — 80% of tasks, good speed/quality
haiku Simple lookups, boilerplate, bulk queries

Quick Reference

bash
# Code review
codex review --uncommitted            # Independent review
codex review --base main              # Review against branch
claude -p "Review uncommitted changes"

# One-shot tasks
codex exec "Fix the failing test"
claude -p "Explain what src/index.ts does"
claude --model haiku -p "Summarize this file"

# Interactive
codex --full-auto                     # Sandboxed, approval on-request
claude
claude --model sonnet

# Resume / continue sessions
codex resume --last                   # Resume last Codex session
claude -c                             # Continue last Claude session
claude -r                             # Pick a Claude session to resume

# Piping
git diff | claude -p "Review this diff"
git diff | codex review -

Tips

  • Second opinion: codex review --uncommitted — different model family eliminates self-bias
  • Model override: codex -m gpt-5.4, claude --model opus/sonnet/haiku. Codex models depend on account type.
  • Cost control: claude -p --max-budget-usd 1.00 caps spending
  • Turn limit: claude -p --max-turns 3 limits agentic loops in print mode

Detailed References

  • Codex CLI
  • Claude Code CLI
  • Gemini CLI

Recommended Agent Skills

Expand your agent's capabilities with these related and highly-rated skills.

vaayne/agent-kit

pi-delegate

Delegate tasks to Pi subagents with preset roles (oracle, worker, reviewer, ui-engineer, librarian) or ad-hoc prompts. Use for fresh context, model specialization, second opinions, code review, isolated execution, or parallel task delegation.

33 3
Explore
vaayne/agent-kit

mcp-jetbrains-ide

Control JetBrains IDE (IntelliJ, WebStorm, PyCharm) via MCP. Use when manipulating IDE files, running configurations, searching code, or performing refactoring. Triggers on "open in IDE", "run configuration", "refactor code", "IDE search", "JetBrains".

33 3
Explore
vaayne/agent-kit

mcp-context7-docs

Query up-to-date documentation and code examples for any programming library or framework. Use when looking up API docs, finding code examples, or checking library usage. Triggers on "how to use [library]", "docs for [package]", "show me examples of [framework]", "Context7 lookup".

33 3
Explore
vaayne/agent-kit

web-fetch

Fetch and extract clean content from URLs using Jina Reader API. Use when users need to read webpage content, extract article text, or fetch URL content for analysis. Triggers on "fetch this page", "read this URL", "extract content from", "get the content of", "what does this page say".

33 3
Explore
vaayne/agent-kit

tmux

Control interactive terminal sessions via tmux. Use when tasks need persistent REPLs, parallel CLI agents, or any process requiring a TTY that simple shell execution cannot handle.

33 3
Explore
vaayne/agent-kit

specs-dev

Plan-first development workflow with review gates. Use when implementing features, refactoring, or any task requiring structured planning, iterative implementation with reviews, and clean commits. Triggers on requests like "implement feature X", "plan and build", "spec-driven development", or when user wants disciplined, reviewed code changes.

33 3
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results