Skip to content

About

Official Deployxa agent plugin for OpenAI Codex - autonomous cloud deployment, streaming build logs, Engine diagnostics, and domain management.

Resources

Code of conduct

Contributing

Security policy

Stars

0 stars

Watchers

0 watching

Forks

Latest commit

 

History

2 Commits

Folders and files

Repository files navigation

Deployxa Agent Plugin for OpenAI Codex (deployxa_codex)

An agentic integration enabling OpenAI Codex (CLI and IDE harness) to autonomously trigger deployments, stream build and runtime logs, run platform diagnostics on failures, and manage cloud infrastructure on the Deployxa platform.


Architecture & MCP Integration

Like Claude Code and ZCode, OpenAI Codex leverages the Model Context Protocol (MCP):

  • Protocol Layer: Exposes Deployxa tools (deployxa_create_deployment, deployxa_get_deployment_logs, deployxa_diagnose_deployment, etc.) via stdio or streamable HTTP.
  • Context & Instructions: Directs Codex through AGENTS.md and domain skills in skills/.
  • Safe Mutations: Write operations (such as creating deployments and updating config) execute directly, while destructive calls (project deletion, domain deletion) require explicit confirmation.

Directory Structure

├── plugin.json               # Codex Agent Plugin manifest
├── AGENTS.md                 # OpenAI Codex agent instruction directives
├── .mcp.json                 # MCP server definitions
├── codex-config.toml         # Configuration template for ~/.codex/config.toml
├── package.json              # Plugin package metadata
├── README.md                 # Setup & usage guide
└── skills/                   # Domain skill instructions
    ├── deployxa-deployment/
    │   └── SKILL.md          # Autonomous build & deployment orchestration
    ├── deployxa-diagnose/
    │   └── SKILL.md          # Failure diagnostics & rollbacks
    └── deployxa-environment/
        └── SKILL.md          # Environment & domain management

Installation & Setup

1. Set Environment Variables

export DEPLOYXA_TOKEN="dxa_your_api_token"
# Optional: only needed if overriding your default organization
# export DEPLOYXA_ORGANIZATION_ID="org_your_org_id"

2. Configure Codex CLI

Add the Deployxa MCP server to Codex via command-line:

# Connect using the hosted Streamable HTTP endpoint (OAuth 2.1 supported):
codex mcp add deployxa --url https://mcp.deployxa.com/mcp

Or append the contents of codex-config.toml to your ~/.codex/config.toml:

[mcp.servers.deployxa]
url = "https://mcp.deployxa.com/mcp"

3. Add AGENTS.md to Your Workspace

Copy or include AGENTS.md in your project root so that the Codex agent harness automatically picks up Deployxa instructions:

curl -o AGENTS.md https://raw.githubusercontent.com/deployxa/codex-plugin/main/AGENTS.md

Example Usage with Codex

Prompt the Codex agent directly:

  • "Deploy this project to Deployxa on the current branch and verify the build status"
  • "Our last deployment failed. Run diagnostics with Deployxa and tell me what broke"
  • "Add domain staging.example.com to my Deployxa project and show required DNS records"
  • "Roll back to deployment d_abc123"

About

Official Deployxa agent plugin for OpenAI Codex - autonomous cloud deployment, streaming build logs, Engine diagnostics, and domain management.

Resources

Code of conduct

Contributing

Security policy

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors