Skip to content

Settings & Configuration

Open the Command Palette (Cmd+Shift+P / Ctrl+Shift+P) and search for “Arcane: Settings”. The settings panel opens in a tab.

ArcaneIDE supports two AI providers. You need at least one API key.

ProviderWhere to get a key
OpenAIplatform.openai.com
Anthropicconsole.anthropic.com

Enter your key in the Settings panel. Keys are stored locally on your machine and never sent anywhere except the provider’s API.

You can select which model powers the AI chat. Available options depend on which API keys you’ve configured.

The default model is chosen automatically based on the current reasoning level. You can pin a specific model with the Explicit Model Override toggle.

By default, ArcaneIDE connects to a local LLM gateway server at http://localhost:3001. If you’re running the Arcane server yourself (e.g. for shared team use), update this URL to point to your instance.

Leave this at the default if you’re using API keys directly.

ActionWhat it does
Reindex ProjectClears the knowledge graph and rebuilds it from scratch
Clear IndexWipes the index without rebuilding (useful for troubleshooting)

Use Reindex Project after large merges or if the AI seems unaware of recent changes.