I bought the unlimitedcodex GPT-5.5 XHigh Weekly or Monthly package.

Help me configure only the local Codex files on this computer for the delivered GPT-5.5 XHigh package family.

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

Safety rules:
1. Back up existing config.toml and auth.json before editing.
2. Never upload, print, log, or share my full API key.
3. If the files do not exist, ask me to open Codex, sign in inside Codex, restart it once, and check the .codex folder again.
4. Use the package model exactly as shown below: gpt-5.5. XHigh is enforced through model_reasoning_effort and the gateway policy.

Use this config.toml:

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

[sandbox_workspace_write]
network_access = true

Use this auth.json shape and replace the placeholder locally:

{
  "auth_mode": "apikey",
  "OPENAI_API_KEY": "YOUR_UNLIMITEDCODEX_API_KEY"
}

Important:
- config.toml base_url must not include /v1.
- Direct API tests use https://claude.kadirr.dev/v1/responses.
- Restart Codex completely after editing.
- Run one small prompt before a long repository task.
- For 401 errors, verify auth.json and the delivered GPT-5.5 XHigh key.
- For 404 errors, remove /v1 from config.toml base_url.