Continue CLI
Secret-free model configuration plus ephemeral environment injection through the secure launcher.
Secret-free model configuration plus ephemeral environment injection through the secure launcher.
| Mode | Protocols | Secret delivery |
|---|---|---|
| Secure launch | openai-responses, openai-chat | Child process environment |
@unlimgpt/cli package.unlimgpt configure continue --dry-run.unlimgpt configure continue.unlimgpt launch continue -- cn --config ~/.continue/config.yaml; configure the IDE extension separately through its native secret UI.unlimgpt launch continue -- cn --config ~/.continue/config.yaml. If you configured --profile work, use unlimgpt launch continue --profile work -- cn --config ~/.continue/config.yaml with your actual profile name. The --config flag opens the local YAML even when a different config was previously selected. In Continue CLI, use /model to select the UnlimGPT model, send “Reply with READY”, wait for a non-empty response, and confirm the request in API Activity. The wording may vary; the request uses credits.unlimgpt keys list, then create a new key with unlimgpt keys create --name rotated-client.Undo: by configure. The CLI leaves later third-party edits intact.| Problem | Resolution |
|---|---|
| The YAML uses anchors or directives. | The CLI stops before writing; use guided setup so YAML semantics are preserved. |
| The IDE cannot read the shell variable. | That is expected for the extension: use Continue's native UI/Hub instead of a plaintext .env. |
If the request fails, run unlimgpt doctor --client continue and see troubleshooting. Diagnostics do not replace a real request.