foxl-ai Codex CLI
Use OpenAI's models directly in Foxl through the Codex CLI. Authenticate with your OpenAI account — Foxl consumes zero relay credits.
Setup
# Install Codex CLI
npm install -g @openai/codex
# or
brew install --cask codex
# Authenticate (choose one)
# Option A: OAuth login
codex login
# Option B: API key
export OPENAI_API_KEY=your-key-here
# Verify
codex exec "hello"
How It Works
Foxl spawns the codex CLI as a subprocess with JSON output. Your OpenAI subscription or API key handles billing — Foxl consumes zero relay credits.
The CLI is invoked with:
codex exec --json -m o4-mini "prompt"
Available Models
| Model ID | Description |
|---|---|
codex-cli/o3 |
o3 — advanced reasoning, complex tasks |
codex-cli/o4-mini |
o4 Mini — fast reasoning, cost-effective |
codex-cli/gpt-4.1 |
GPT 4.1 — 1M context, general purpose |
Credential Detection
Foxl checks these sources (in order):
- Environment variable:
OPENAI_API_KEY - Auth file:
~/.codex/auth.json - macOS Keychain: service "Codex Auth"
Features
- JSONL streaming (via
exec --json) - Reasoning model support (o3, o4-mini)
- Tool call events in stream
- Auto-detect credentials on startup
Troubleshooting
"Codex CLI not configured"
Run codex login or set OPENAI_API_KEY environment variable.
401 Unauthorized
Your OpenAI API key may be invalid or expired. Run codex login again or check your key at platform.openai.com.
Model not found
Some models require specific API tier access. Check your OpenAI account for available models.