microsoft/autogen · error · ValueError

endpoint must be a valid URL starting with http:// or https:

Error message

endpoint must be a valid URL starting with http:// or https://

What it means

AzureAISearchConfig is a pydantic model whose endpoint field validator rejects any value not starting with http:// or https://. The search SDK needs a full URL, so a bare hostname fails fast at config construction (which _validate_config surfaces as 'Invalid configuration: ...', error 925).

Source

Thrown at python/packages/autogen-ext/src/autogen_ext/tools/azure/_config.py:148

    enable_caching: bool = Field(default=False, description="Whether to cache search results")
    cache_ttl_seconds: int = Field(default=300, description="How long to cache results in seconds")

    embedding_provider: Optional[str] = Field(
        default=None, description="Name of embedding provider for client-side embeddings"
    )
    embedding_model: Optional[str] = Field(default=None, description="Model name for client-side embeddings")
    openai_api_key: Optional[str] = Field(default=None, description="API key for OpenAI/Azure OpenAI embeddings")
    openai_api_version: Optional[str] = Field(default=None, description="API version for Azure OpenAI embeddings")
    openai_endpoint: Optional[str] = Field(default=None, description="Endpoint URL for Azure OpenAI embeddings")

    model_config = {"arbitrary_types_allowed": True}

    @field_validator("endpoint")
    def validate_endpoint(cls, v: str) -> str:
        """Validate that the endpoint is a valid URL."""
        if not v.startswith(("http://", "https://")):
            raise ValueError("endpoint must be a valid URL starting with http:// or https://")
        return v

    @field_validator("query_type")
    def normalize_query_type(cls, v: QueryTypeLiteral) -> QueryTypeLiteral:
        """Normalize query type to standard values."""
        if not v:
            return "simple"

        if isinstance(v, str) and v.lower() == "fulltext":
            return "full"

        return v

    @field_validator("top")
    def validate_top(cls, v: Optional[int]) -> Optional[int]:
        """Ensure top is a positive integer if provided."""
        if v is not None and v <= 0:
            raise ValueError("top must be a positive integer")

View on GitHub (pinned to 027ecf0a37)

Solutions

1. Use the full endpoint: https://<service-name>.search.windows.net.
2. If you only have the service name, build the endpoint programmatically: f'https://{name}.search.windows.net'.
3. Validate/normalize endpoints in config loading (strip whitespace, prepend scheme) before constructing the tool.

Example fix

# before
endpoint=os.environ['AZURE_SEARCH_SERVICE']  # 'my-svc'
# after
endpoint=f"https://{os.environ['AZURE_SEARCH_SERVICE']}.search.windows.net"

def endpoint_is_valid(endpoint) -> bool:
    return isinstance(endpoint, str) and endpoint.strip().startswith(('http://', 'https://'))

def is_search_endpoint(endpoint) -> bool:
    import re
    return bool(re.match(r'^https?://[\w.-]+\.search\.windows\.net/?$', endpoint or ''))

When it happens

Trigger: Passing endpoint='my-svc.search.windows.net' (no scheme), endpoint='my-svc', or a value with leading whitespace like ' https://...'.

Common situations: Copying the search service name from the portal instead of the URL; env var storing just the resource name; configuration files that trim or mangle the scheme; typos like 'https:/svc' (single slash).

Related errors

• Invalid configuration: {str(e)}
• top must be a positive integer
• semantic_config_name must be provided when query_type is 'se
• vector_fields must be provided for vector search
• vector_fields must contain at least one field name for vecto

AI-assisted analysis of microsoft/autogen@027ecf0a37 (2026-08-15). Data as JSON: /api/errors/9d60bedbc2c7fb41. Report an issue: GitHub.