I want to configure Codex CLI / Codex IDE with my unlimitedcodex API key.

Please help me update only my local Codex configuration files on this computer.

Files:
- Windows config: %USERPROFILE%\.codex\config.toml
- Windows auth: %USERPROFILE%\.codex\auth.json
- macOS/Linux config: ~/.codex/config.toml
- macOS/Linux auth: ~/.codex/auth.json

Important:
- If config.toml or auth.json does not exist, do not guess a random location.
- First ask me to open Codex, sign in to ChatGPT inside Codex, close and reopen Codex once, then check the .codex folder again.
- In config.toml, base_url must be exactly https://claude.kadirr.dev with no /v1 suffix.
- Use /v1 only for direct API requests, for example https://claude.kadirr.dev/v1/responses.
- Recommended current model: gpt-5.5.
- 5.6 model choices are temporarily paused; use gpt-5.5 until support says otherwise.
- Active compatibility model: gpt-5.5. Use this while 5.6 models are paused.

Before editing:
1. Back up any existing config.toml and auth.json files.
2. Do not upload, print, log, or share my full API key.
3. Do not put my API key into any repository, screenshot, public chat, browser frontend code, or shared document.

Use this config.toml content unless I explicitly choose another model:

model_provider = "OpenAI"
model = "gpt-5.5"
review_model = "gpt-5.5"
model_reasoning_effort = "xhigh"
disable_response_storage = true
network_access = "enabled"

[model_providers.OpenAI]
name = "OpenAI"
base_url = "https://claude.kadirr.dev"
wire_api = "responses"
requires_openai_auth = true

[features]
goals = true

If support explicitly approves another model later, change only these two lines:

model = "gpt-5.5"
review_model = "gpt-5.5"

Do not use gpt-5.6-sol right now. The default new setup should stay gpt-5.5.

Use this auth.json shape. Replace YOUR_UNLIMITEDCODEX_API_KEY with the API key I provide:

{
  "OPENAI_API_KEY": "YOUR_UNLIMITEDCODEX_API_KEY"
}

After editing:
1. Ask me to restart Codex CLI / Codex IDE / Codex app.
2. Ask me to run one small test prompt first.
3. If there is an error, help me read the error without exposing my full API key.
4. If I get 401 or invalid_api_key, check the key and auth.json field.
5. If I get 404 or route not found, check that config.toml base_url does not contain /v1.