codex

name	codex
description	Delegate coding to OpenAI Codex CLI (features, PRs).
version	1.0.0
author	Hermes Agent
license	MIT
platforms	["linux","macos","windows"]
metadata	{"hermes":{"tags":["Coding-Agent","Codex","OpenAI","Code-Review","Refactoring"],"related_skills":["claude-code","hermes-agent"]}}

Codex CLI

Delegate coding tasks to Codex via the Hermes terminal. Codex is OpenAI's autonomous coding agent CLI.

When to use

• Building features
• Refactoring
• PR reviews
• Batch issue fixing

Requires the codex CLI and a git repository.

Prerequisites

• Codex installed (two methods):
  • npm: npm install -g @openai/codex
  • Standalone binary (Rust, recommended): download from GitHub Releases
    • Linux x86_64: codex-x86_64-unknown-linux-musl.tar.gz
    • Linux ARM: codex-aarch64-unknown-linux-musl.tar.gz
    • macOS ARM: codex-aarch64-apple-darwin.tar.gz
    • macOS x86: codex-x86_64-apple-darwin.tar.gz
    • Windows x86: codex-x86_64-pc-windows-msvc.exe.zip
    • Version naming: rust-v0.139.0 format (the rust- prefix indicates Rust build)
  • Extract and place binary in PATH
• OpenAI auth configured: either OPENAI_API_KEY or Codex OAuth credentials from the Codex CLI login flow
• Must run inside a git repository — Codex refuses to run outside one
• Network: needs access to OpenAI API (api.openai.com) — requires proxy on servers without direct access
• Use pty=true in terminal calls — Codex is an interactive terminal app

For Hermes itself, model.provider: openai-codex uses Hermes-managed Codex OAuth from ~/.hermes/auth.json after hermes auth add openai-codex. For the standalone Codex CLI, a valid CLI OAuth session may live under ~/.codex/auth.json; do not treat a missing OPENAI_API_KEY alone as proof that Codex auth is missing.

One-Shot Tasks

terminal(command="codex exec 'Add dark mode toggle to settings'", workdir="~/project", pty=true)

For scratch work (Codex needs a git repo):

terminal(command="cd $(mktemp -d) && git init && codex exec 'Build a snake game in Python'", pty=true)

Background Mode (Long Tasks)

# Start in background with PTY
terminal(command="codex exec --full-auto 'Refactor the auth module'", workdir="~/project", background=true, pty=true)
# Returns session_id

# Monitor progress
process(action="poll", session_id="<id>")
process(action="log", session_id="<id>")

# Send input if Codex asks a question
process(action="submit", session_id="<id>", data="yes")

# Kill if needed
process(action="kill", session_id="<id>")

Key Flags

Flag	Effect
exec "prompt"	One-shot execution, exits when done
--full-auto	Sandboxed but auto-approves file changes in workspace
--yolo	No sandbox, no approvals (fastest, most dangerous)

PR Reviews

Clone to a temp directory for safe review:

terminal(command="REVIEW=$(mktemp -d) && git clone https://github.com/user/repo.git $REVIEW && cd $REVIEW && gh pr checkout 42 && codex review --base origin/main", pty=true)

Parallel Issue Fixing with Worktrees

# Create worktrees
terminal(command="git worktree add -b fix/issue-78 /tmp/issue-78 main", workdir="~/project")
terminal(command="git worktree add -b fix/issue-99 /tmp/issue-99 main", workdir="~/project")

# Launch Codex in each
terminal(command="codex --yolo exec 'Fix issue #78: <description>. Commit when done.'", workdir="/tmp/issue-78", background=true, pty=true)
terminal(command="codex --yolo exec 'Fix issue #99: <description>. Commit when done.'", workdir="/tmp/issue-99", background=true, pty=true)

# Monitor
process(action="list")

# After completion, push and create PRs
terminal(command="cd /tmp/issue-78 && git push -u origin fix/issue-78")
terminal(command="gh pr create --repo user/repo --head fix/issue-78 --title 'fix: ...' --body '...'")

# Cleanup
terminal(command="git worktree remove /tmp/issue-78", workdir="~/project")

Batch PR Reviews

# Fetch all PR refs
terminal(command="git fetch origin '+refs/pull/*/head:refs/remotes/origin/pr/*'", workdir="~/project")

# Review multiple PRs in parallel
terminal(command="codex exec 'Review PR #86. git diff origin/main...origin/pr/86'", workdir="~/project", background=true, pty=true)
terminal(command="codex exec 'Review PR #87. git diff origin/main...origin/pr/87'", workdir="~/project", background=true, pty=true)

# Post results
terminal(command="gh pr comment 86 --body '<review>'", workdir="~/project")

Server Compatibility Check

When user asks "can server X run Codex", check:

uname -m && free -h && df -h / && cat /etc/os-release | head -3

Minimum: x86_64 or aarch64, ~200MB free RAM, ~100MB disk. The binary itself is lightweight — main resource cost is the API call context, not local compute.

Key pitfall: Codex needs outbound HTTPS to api.openai.com. Servers in China (Alibaba Cloud etc.) without a proxy will fail silently or timeout. Always check network access before recommending installation.

Custom Providers (Non-OpenAI)

Codex supports non-OpenAI providers via config fields discovered in the config schema:

Config Key	Purpose
model_provider	Select active provider from the model_providers map
model_providers	Define custom provider entries (extends built-in list)
openai_base_url	Override the built-in openai provider's base URL
model	Model name to use

Quick setup for OpenAI-compatible APIs (DeepSeek, Together AI, etc.):

# In ~/.codex/config.json or via environment:
export OPENAI_BASE_URL="https://api.deepseek.com/v1"
export OPENAI_API_KEY="sk-..."
export MODEL="deepseek-coder"

Works with: DeepSeek, Together AI, SiliconFlow, Ollama (local), vLLM, any OpenAI-compatible endpoint.

Caveat: Codex requires strong tool-calling and code-understanding capabilities. Smaller/weaker models will produce poor results. GPT-4-class or equivalent recommended.

Pitfalls

1. Always use pty=true — Codex is an interactive terminal app and hangs without a PTY
2. Git repo required — Codex won't run outside a git directory. Use mktemp -d && git init for scratch
3. Large binary downloads — Standalone binaries are 60-100MB. On bandwidth-limited servers, ask user before downloading. Consider npm install instead if Node.js is available.
4. China servers — api.openai.com is blocked. Need proxy or alternative endpoint.
5. --full-auto for building — auto-approves changes within the sandbox
6. Background for long tasks — use background=true and monitor with process tool
7. Don't interfere — monitor with poll/log, be patient with long-running tasks
8. Parallel is fine — run multiple Codex processes at once for batch work
9. Cancel cleanup — if a large binary download is interrupted, rm partial files immediately (codex-*.tar.gz, codex-*.zip)