One-click install
npx skills add https://github.com/dustland/openviber --skill codex-cli-dustland
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: codex-cli
Source: https://github.com/dustland/openviber/tree/main/src/skills/codex-cli
Command: npx skills add https://github.com/dustland/openviber --skill codex-cli-dustland

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates autonomous coding tasks by running the OpenAI Codex CLI from within Viber, enabling hands-free coding workflows directly from chat.

Core Features & Use Cases

  • Runs Codex CLI non-interactively to perform coding tasks, analyses, and small refactors with chat-friendly outputs.
  • Returns a structured result including status, exitCode, stdoutTail, and stderrTail, designed for UI-friendly display.
  • Supports multiple execution modes (full-auto, workspace-write, read-only) and optional Codex model overrides for flexibility across local projects.

Quick Start

Ask Codex to perform a coding task by providing a prompt to codex_run.

Frequently Asked Questions about codex-cli

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

FAQPage Schema
How do I run autonomous coding tasks from a chat interface?

You can run autonomous coding tasks from a chat interface by invoking the Codex CLI non-interactively. This enables hands-free local software development workflows, executing code generation, analysis, and refactoring directly from your chat application.

What execution modes are available for local code generation and refactoring?

Local code generation and refactoring support three execution modes: full-auto, workspace-write, and read-only. These modes control the level of file system access and autonomy during the coding task.

Can I override the AI model used for code analysis in my local development environment?

Yes, you can override the AI model used for code analysis in your local development environment. The execution supports optional Codex model overrides, providing flexibility across different local projects.

How does autonomous coding return results for UI-friendly display?

Autonomous coding returns structured results designed for UI-friendly display by including the execution status, exitCode, stdoutTail, and stderrTail. This structured format ensures outputs are chat-friendly and easy to read.

Does the Codex CLI Skill enforce environment checks before task automation?

Yes, the Codex CLI Skill enforces environment checks before task automation. This ensures the local development setup meets the necessary prerequisites before executing any autonomous coding commands.

What is the best way to automate small code refactors without manual intervention?

The best way to automate small code refactors without manual intervention is using full-auto mode via the Codex CLI. It performs non-interactive coding tasks and returns structured results including exit codes and output tails.