No available model to generate agent specification.

Error message

No available model to generate agent specification.

What it means

After resolving model patterns against the registry, the architect picks the resolved model or falls back to the first available model. If the registry has no available models at all, there is no LLM to run the architect prompt, so this error is thrown.

Source

Thrown at packages/coding-agent/src/modes/components/agents-hub.ts:762

		}
	}

	async #runAgentCreationArchitect(description: string): Promise<GeneratedAgentSpec> {
		const modelRegistry = this.#modelContext.modelRegistry;
		if (!modelRegistry) {
			throw new Error("Model registry unavailable in current session.");
		}
		await modelRegistry.refresh();
		const modelPatterns = resolveConfiguredModelPatterns(
			this.#modelContext.activeModelPattern ??
				this.#modelContext.defaultModelPattern ??
				this.#settings.getModelRole("default"),
			this.#settings,
		);
		const { model } = resolveModelOverride(modelPatterns, modelRegistry, this.#settings);
		const selectedModel = model ?? modelRegistry.getAvailable()[0];
		if (!selectedModel) {
			throw new Error("No available model to generate agent specification.");
		}
		const systemPrompt = prompt.render(agentCreationArchitectPrompt, {});
		const userPrompt = prompt.render(agentCreationUserPrompt, { request: description });
		const { session } = await createAgentSession({
			cwd: this.#cwd,
			authStorage: modelRegistry.authStorage,
			modelRegistry,
			settings: this.#settings,
			model: selectedModel,
			systemPrompt: [systemPrompt],
			hasUI: false,
			enableLsp: false,
			enableMCP: false,
			disableExtensionDiscovery: true,
			toolNames: ["__none__"],
			customTools: [],
			skills: [],
			contextFiles: [],

View on GitHub (pinned to 9690622007)

Solutions

Log in / configure API keys (e.g. run auth setup) so at least one model is available
Clear or widen the configured model patterns (settings default role / activeModelPattern) so they match an existing model
Run modelRegistry.refresh() or restart the session to re-detect providers
Check network/provider status and retry

Example fix

// before
settings default role: "gpt-9-nonexistent" // matches nothing
// after
settings default role: "claude-sonnet-4" // available model

Defensive patterns

Strategy: validation

Validate before calling

const available = modelRegistry.getAvailable();
if (available.length === 0) {
  throw new Error("No models available — check auth and provider config before creating agents");
}

Type guard

function hasAvailableModels(reg: { getAvailable(): unknown[] }): boolean {
  return reg.getAvailable().length > 0;
}

Try / catch

try {
  await hub.runAgentCreationArchitect(desc);
} catch (err) {
  if (err instanceof Error && err.message.startsWith("No available model")) {
    // prompt user to configure API keys / fix model patterns
  } else throw err;
}

Prevention

Configure auth/API keys before using LLM-dependent features
Keep the default model role pointing at a model that exists in the registry
Avoid over-restrictive model patterns that filter out everything
Call registry.refresh() after changing provider configuration

When it happens

Trigger: modelRegistry.getAvailable() returns an empty array and resolveModelOverride returned no concrete model — e.g. no auth credentials, all providers unreachable, or pattern filters exclude every model.

Common situations: Fresh install without API keys logged in; provider outage or revoked API key; configured model patterns (default role / active pattern) matching zero models in the registry.

Related errors

AI-assisted analysis of can1357/oh-my-pi@9690622007 (2026-08-31). Data as JSON: /api/errors/4df9ed79643478f2. Report an issue: GitHub.