microsoft/autogen · error · Error

Template ${templateId} not found for component type ${compon

Error message

Template ${templateId} not found for component type ${componentType}

What it means

Raised on the azure_openai embedding path when `openai_endpoint` is missing from the config. Azure OpenAI clients are constructed per-resource (endpoint-scoped), unlike plain OpenAI, so a resource URL is mandatory; api_key can be omitted in favor of DefaultAzureCredential but the endpoint cannot.

Source

Thrown at python/packages/autogen-studio/frontend/src/components/types/component-templates.ts:530

  const templates = COMPONENT_TEMPLATES[componentType];
  return templates.find((template) => template.id === templateId);
}

export function getDefaultTemplate(
  componentType: ComponentTypes
): ComponentTemplate<ComponentConfig> | undefined {
  const templates = COMPONENT_TEMPLATES[componentType];
  return templates[0]; // Return first template as default
}

export function createComponentFromTemplate(
  templateId: string,
  componentType: ComponentTypes,
  overrides?: Partial<Component<ComponentConfig>>
): Component<ComponentConfig> {
  const template = getTemplateById(componentType, templateId);
  if (!template) {
    throw new Error(
      `Template ${templateId} not found for component type ${componentType}`
    );
  }

  return {
    provider: template.provider,
    component_type: template.component_type,
    version: template.version,
    component_version: template.component_version,
    description: template.description,
    config: template.config,
    label: template.label,
    ...overrides,
  };
}

// Workbench-specific helper functions
export interface WorkbenchDropdownOption {

View on GitHub (pinned to 027ecf0a37)

Solutions

Set `openai_endpoint` to your Azure OpenAI resource URL, e.g. "https://<resource-name>.openai.azure.com".
Confirm you are not accidentally reusing the AI Search endpoint value; they are different resources.
If you meant to use non-Azure OpenAI, set embedding_provider="openai" instead (no endpoint needed).

Example fix

# before
config = AzureAISearchConfig(
    ..., embedding_provider="azure_openai", embedding_model="text-embedding-3-large",
)  # missing endpoint

# after
config = AzureAISearchConfig(
    ...,
    embedding_provider="azure_openai",
    embedding_model="text-embedding-3-large",
    openai_endpoint="https://my-aoai.openai.azure.com",
    openai_api_version="2024-02-01",
)

Defensive patterns

Strategy: validation

Validate before calling

def validate_azure_openai(cfg) -> None:
    if str(cfg.embedding_provider or "").lower() == "azure_openai" and not getattr(cfg, "openai_endpoint", None):
        raise ValueError("openai_endpoint is required for azure_openai embeddings")

Prevention

Keep AI Search endpoint and OpenAI endpoint as separate, clearly named env vars.
Use pydantic validation on your own settings model to require the endpoint when provider is azure_openai.

When it happens

Trigger: Configuring embedding_provider='azure_openai' and embedding_model but not setting `openai_endpoint` (e.g. only openai_api_key set, or copying a plain-OpenAI config and switching provider), then running a vector search.

Common situations: Confusing the Azure AI Search endpoint (which IS set) with the Azure OpenAI endpoint (a separate resource); forgetting the endpoint when using key-based auth because examples show key-only for plain OpenAI; environment-specific config where the endpoint env var is unset in one environment.

Related errors

AI-assisted analysis of microsoft/autogen@027ecf0a37 (2026-08-15). Data as JSON: /api/errors/d5ec04a6dda2f391. Report an issue: GitHub.