microsoft/semantic-kernel · critical · ServiceInitializationError

Failed to create Anthropic settings.

Error message

Failed to create Anthropic settings.

What it means

During AnthropicChatCompletion.__init__, AnthropicSettings is constructed from the api_key, model id, env file, and environment. If pydantic validation fails (most commonly a missing or malformed API key), the ValidationError is wrapped as ServiceInitializationError and chained as __cause__.

Source

Thrown at python/semantic_kernel/connectors/ai/anthropic/services/anthropic_chat_completion.py:110

            ai_model_id: Anthropic model name, see
                https://docs.anthropic.com/en/docs/about-claude/models#model-names
            service_id: Service ID tied to the execution settings.
            api_key: The optional API key to use. If provided will override,
                the env vars or .env file value.
            async_client: An existing client to use.
            env_file_path: Use the environment settings file as a fallback
                to environment variables.
            env_file_encoding: The encoding of the environment settings file.
        """
        try:
            anthropic_settings = AnthropicSettings(
                api_key=api_key,
                chat_model_id=ai_model_id,
                env_file_path=env_file_path,
                env_file_encoding=env_file_encoding,
            )
        except ValidationError as ex:
            raise ServiceInitializationError("Failed to create Anthropic settings.", ex) from ex

        if not anthropic_settings.chat_model_id:
            raise ServiceInitializationError("The Anthropic chat model ID is required.")

        if not async_client:
            async_client = AsyncAnthropic(
                api_key=anthropic_settings.api_key.get_secret_value(),
            )

        super().__init__(
            async_client=async_client,
            service_id=service_id or anthropic_settings.chat_model_id,
            ai_model_id=anthropic_settings.chat_model_id,
        )

    # region Overriding base class methods

    # Override from AIServiceClientBase

View on GitHub (pinned to c028a0c7dc)

Solutions

1. Set ANTHROPIC_API_KEY in the environment or .env, or pass api_key explicitly to the constructor.
2. Inspect the chained ValidationError (`e.__cause__`) to see which field(s) failed.
3. Verify env_file_path and env_file_encoding point to a readable file if you rely on .env fallback.

Example fix

// before
svc = AnthropicChatCompletion()  # no ANTHROPIC_API_KEY anywhere -> ValidationError wrapped

// after
svc = AnthropicChatCompletion(
    ai_model_id="claude-3-5-sonnet-latest",
    api_key=os.environ["ANTHROPIC_API_KEY"],
)

import os
key = os.environ.get('ANTHROPIC_API_KEY') or api_key
if not key or len(key) < 20:
    raise ValueError('ANTHROPIC_API_KEY missing or malformed')

def has_valid_anthropic_key() -> bool:
    key = os.environ.get('ANTHROPIC_API_KEY')
    return bool(key) and len(key) >= 20

from semantic_kernel.exceptions.service_exceptions import ServiceInitializationError
try:
    svc = AnthropicChatCompletion(ai_model_id=model, api_key=key)
except ServiceInitializationError as e:
    cause = e.__cause__  # pydantic ValidationError with field details
    raise SystemExit(f'Anthropic init failed: {cause}')

When it happens

Trigger: Constructing AnthropicChatCompletion without a valid ANTHROPIC_API_KEY (not in env, not in .env, not passed), with a malformed key, or with an unreadable/mis-encoded env file.

Common situations: No ANTHROPIC_API_KEY in the environment; wrong env var name; .env file path or encoding mismatch; key passed as an empty string; secrets not loaded in the deployment.

Related errors

• The Anthropic chat model ID is required.
• Failed to create Google settings.
• Failed to create MongoDB Atlas settings.
• Failed to create Azure OpenAI settings: {exc}
• Please provide an Azure OpenAI endpoint

AI-assisted analysis of microsoft/semantic-kernel@c028a0c7dc (2026-08-13). Data as JSON: /api/errors/3aebba996d9bc0da. Report an issue: GitHub.