codeagent

Automate multi-backend AI code tasks across Codex, Claude, and Gemini backends.

4|2|Updated Jan 16, 2026
One-click install
npx skills add https://github.com/PeterFile/my-common-skills --skill codeagent-peterfile
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codeagent
Source: https://github.com/PeterFile/my-common-skills/tree/main/codeagent
Command: npx skills add https://github.com/PeterFile/my-common-skills --skill codeagent-peterfile

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill wraps the codeagent-wrapper to coordinate multi-backend AI code tasks (Codex, Claude, Gemini), enabling file-referenced inputs and structured outputs across backends.

Core Features & Use Cases

  • Backends & Orchestration: Run tasks on Codex, Claude, and Gemini with per-task backend selection and parallel execution.
  • File References: Use @syntax to reference code files and drive deterministic tasks.
  • Structured Outputs: Produce uniform results ready for review, integration, or automation.
  • Use Case: Orchestrate a large refactor across multiple files and languages, coordinating code analysis, generation, and review in a single workflow.

Quick Start

Use codeagent-wrapper to execute a simple code task in a working directory: codeagent-wrapper - [working_dir] <<'EOF' <task content here> EOF

Frequently Asked Questions about codeagent

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

FAQPage Schema
How do I orchestrate AI code tasks across multiple backends like Codex and Claude?

Multi-backend AI code task orchestration coordinates Codex, Claude, and Gemini to execute parallel tasks. This Skill wraps the codeagent-wrapper to enable per-task backend selection and returns structured outputs for automated workflows.

Can I reference specific code files when running automated code generation?

Automated code generation supports file references using @syntax. You provide inputs referencing specific code files within the codeagent-wrapper, enabling deterministic task execution and structured outputs suitable for follow-up steps.

What is the best way to run a large-scale refactoring workflow across different languages?

Large-scale refactoring workflows coordinate code analysis, generation, and review across multiple files and languages. Using multi-agent orchestration, you select backends per task and execute them in parallel within a single workflow.

Do I need the codeagent-wrapper binary to use multi-agent code orchestration?

Multi-agent code orchestration requires the codeagent-wrapper binary. You execute tasks by passing a working directory and task content to the binary, which handles the backend dispatch and returns uniform structured results.

Does multi-backend AI coding support parallel execution for complex code analysis?

Complex code analysis supports parallel execution across Codex, Claude, and Gemini backends. This multi-backend orchestration allows you to process multiple tasks simultaneously, producing structured outputs ready for integration or automated review.