GatewayCoding Agents

Codex CLI (gateway)

Route Codex CLI’s LLM calls through the Respan gateway for automatic logging, caching, and model switching after Responses API access is configured for your Respan key.

Setup

1

Install Codex CLI

$npm install -g @openai/codex
2

Set environment variable

$export RESPAN_API_KEY="YOUR_RESPAN_API_KEY"

No OPENAI_API_KEY is needed in your shell, but the Responses API route must be enabled for your Respan key.

3

Configure Codex

Add to ~/.codex/config.toml (user-level) or .codex/config.toml (project-level):

1model = "gpt-5.5"
2model_provider = "respan"
3
4[model_providers.respan]
5name = "Respan Gateway"
6base_url = "https://api.respan.ai/api/"
7wire_api = "responses"
8env_key = "RESPAN_API_KEY"

env_key is the name of the environment variable that holds your API key, not the actual key. Current Codex CLI requires wire_api = "responses"; wire_api = "chat" is rejected by Codex itself. Live verification with a standard Respan gateway key returned a provider-credential error on /api/responses, so use this setup only after Responses API access is configured for your key.

4

Run Codex

$codex "your prompt"

All LLM calls now route through Respan.

Switch models

Change the model parameter in ~/.codex/config.toml to use any supported model.

1model = "gpt-5.5"
2# model = "claude-sonnet-4-5-20250929"
3# model = "gemini/gemini-3.5-flash"

See the full model list.