Reference
Arcane.ai
Provider-neutral AI profile, model, chat, and provider-setting operations.
This focused page is derived from the mechanically checked full member inventory.
Syntax
Arcane.ai
Member kind
Namespace
Description
Provider-neutral AI profile, model, chat, and provider-setting operations.
Overview
Arcane.ai is the provider-neutral surface for model inventory, effective AI
profile, chat, provider settings, and account-visible provider models. It keeps
application code above raw provider protocols. Request and result objects use
the checked contracts linked from the canonical method pages.
Availability and security
Inference requires ai.inference. Raw local inventory is restricted to
Settings, Shell, and Terminal with ai.models.read; provider settings and model
discovery are Settings-only with ai.settings.manage. Credentials stay inside
the native settings boundary and must not be logged or copied into prompts.
Example
const access = await globalThis.Arcane.capabilities.list();
const canChat = access.methods.includes('ai.chat');
console.log('Provider-neutral chat admitted:', canChat);
Members
Arcane.ai.models()— Lists raw local model inventory for Settings, Terminal, and Shell diagnostics. Application model selectors use the admitted Arcane.localAI.status().models.ollama catalog instead.Arcane.ai.chat()— Sends a chat request through the configured provider.Arcane.ai.profile()— Gets the effective AI profile, including its canonical conversational response-length target.Arcane.ai.providerSettings()— Gets provider selection, credential status, and the conversational response-length target.Arcane.ai.saveProviderSettings()— Saves provider settings; validates the selected model when OpenAI is selected and validates a supplied replacement credential before persistence. Ollama-selected settings may retain a dormant OpenAI model for later revalidation.Arcane.ai.providerModels()— Queries models available to the configured OpenAI account.