apache/seatunnel · error · ImportError

anthropic package required for AI_PROVIDER=anthropic…

Error message

anthropic package required for AI_PROVIDER=anthropic. Install it: pip install anthropic

What it means

AnthropicProvider initializes the Anthropic Messages API client and imports the third-party 'anthropic' package lazily. If the package is not installed, it raises ImportError with an install hint before any API call is made.

Solutions

Run: pip install anthropic in the active environment
Install with the CLI's LLM extra if provided (e.g. pip install 'seatunnel-cli[anthropic]')
Verify with 'python -c "import anthropic"' that the same interpreter running the CLI has the package

Example fix

// before
export AI_PROVIDER=anthropic
seatunnel-cli ...   # ImportError
// after
pip install anthropic
export AI_PROVIDER=anthropic

Defensive patterns

Strategy: try-catch

Validate before calling

python -c "import anthropic" || pip install anthropic

Try / catch

try:
    provider = get_provider()
except ImportError as e:
    print(f'Missing LLM dependency: {e}. Run: pip install anthropic')
    sys.exit(2)

Prevention

Install CLI LLM extras in the same venv used to run the CLI
Pin anthropic in your requirements/setup
Verify imports in CI before invoking LLM features
Avoid switching AI_PROVIDER without installing its package

When it happens

Trigger: AI_PROVIDER=anthropic selects AnthropicProvider and its __init__ runs 'import anthropic', which fails because the package is absent from the current Python environment.

Common situations: Installed seatunnel-cli without the anthropic extra; running inside a fresh venv/container lacking the dependency; forgot 'pip install anthropic' after switching AI_PROVIDER.

Understand the failure class

Background: "X is not installed. Please install it with pip install Y": missing optional dependency errors — ImportError/ValueError raised when a library's optional extra was never installed — this error's family across 22 libraries.

Related errors

AI-assisted analysis of apache/seatunnel@cf67b549a7 (2026-09-10). Data as JSON: /api/errors/9aa4065af84f06a6. Report an issue: GitHub.

Appendix: source

Thrown at seatunnel-cli/seatunnel_cli/llm_provider.py:478

                    yield {"type": "tool_input_delta", "tool_use_id": current_tool_id or "", "delta": delta["toolUse"].get("input", "")}
            elif "contentBlockStop" in event:
                if current_tool_id:
                    yield {"type": "tool_stop", "tool_use_id": current_tool_id}
                    current_tool_id = None
            elif "messageStop" in event:
                yield {"type": "message_stop", "stop_reason": event["messageStop"].get("stopReason", "end_turn")}


# ─── Anthropic Provider ───

class AnthropicProvider(LLMProvider):
    """Anthropic Messages API provider (direct, not via Bedrock)."""

    def __init__(self):
        try:
            import anthropic
        except ImportError:
            raise ImportError(
                "anthropic package required for AI_PROVIDER=anthropic. "
                "Install it: pip install anthropic"
            )

        api_key = os.environ.get("ANTHROPIC_API_KEY")
        if not api_key:
            raise ValueError("ANTHROPIC_API_KEY environment variable is required for AI_PROVIDER=anthropic")

        self._client = anthropic.Anthropic(api_key=api_key)
        self._model_id = os.environ.get("ANTHROPIC_MODEL", "claude-sonnet-4-20250514")
        self._fast_model_id = os.environ.get("ANTHROPIC_SMALL_FAST_MODEL", "claude-haiku-4-5-20251001")

    @property
    def provider_name(self) -> str:
        return "anthropic"

    @property
    def model_id(self) -> str:

View on GitHub (pinned to cf67b549a7)