# Continue

Continue can use Codex Pooler as an OpenAI-compatible provider by setting `provider: openai`, `apiBase` to `/v1`, and the Pool API key as a Continue secret. For `gpt-5*` models, Continue uses the Responses API by default.

## Config file path

For local Continue configs, put the Codex Pooler assistant in `config.yaml`:

| OS | Local config file |
| --- | --- |
| macOS | `~/.continue/config.yaml` |
| Linux | `~/.continue/config.yaml` |
| Windows | `%USERPROFILE%\.continue\config.yaml` |

In the IDE extension, open the Continue chat sidebar, use the config selector above the chat input, then click the gear icon beside **Local Config**. Continue creates this file on first use and reloads it after saves.

Continue CLI resolves config from `--config` first, then the saved last-used config, then the default assistant or `~/.continue/config.yaml` when not logged in.

Hub configs live in Continue Mission Control instead of this local file. For Codex Pooler examples, local `config.yaml` is the clearest setup because it keeps the Pool route, capabilities, and optional MCP entry inspectable.

```yaml
name: Codex Pooler
version: 1.0.0
schema: v1

models:
  - name: GPT-5.5 via Codex Pooler
    provider: openai
    model: gpt-5.5
    apiBase: https://codex-pooler.example.com/v1
    apiKey: "${{ secrets.CODEX_POOLER_API_KEY }}"
    contextLength: 272000
    defaultCompletionOptions:
      maxTokens: 128000
    roles:
      - chat
      - edit
      - apply
      - summarize
    capabilities:
      - tool_use
      - image_input

# Optional operator-only MCP metadata add-on. Omit for model/runtime use.
mcpServers:
  - name: codex_pooler
    type: streamable-http
    url: https://codex-pooler.example.com/mcp
    requestOptions:
      timeout: 30000
      headers:
        Authorization: "Bearer ${{ secrets.CODEX_POOLER_MCP_KEY }}"
```

For local setup, change `apiBase` to `http://localhost:4000/v1`. If you keep the optional operator MCP add-on locally, change the MCP `url` to `http://localhost:4000/mcp`.

Continue uses `contextLength` for request pruning and `defaultCompletionOptions.maxTokens` for the completion budget. It prunes request messages before sending instead of doing Codex-style local compaction, so keep the context length at Codex Pooler's exposed model window.

Check the headless CLI path after saving the config:

```bash
export CODEX_POOLER_API_KEY=<pool-api-key>
npx -y @continuedev/cli@latest -p \
  --config ~/.continue/config.yaml \
  --silent \
  'Reply with exactly: continue ok'
```

The Pool API key authenticates model requests. The MCP token authenticates only the operator metadata endpoint.