What causes "Expected Msg object, got {type(msg)} instead." in agentscope-ai/agentscope?

Deserialize dicts back into Msg objects before formatting (e.g. Msg.from_dict(d))

agentscope-ai/agentscope · error · TypeError

Expected Msg object, got

Error message

Expected Msg object, got {type(msg)} instead.

What it means

assert_list_of_msgs iterates the input list and raises TypeError when any element is not a Msg instance. This catches mixed or converted histories, e.g. dicts from JSON deserialization or plain strings, before the formatter chokes deeper in processing.

Solutions

Deserialize dicts back into Msg objects before formatting (e.g. Msg.from_dict(d))
Filter or map the list so every element is a Msg
Fix the producer that inserts strings/dicts into the history

Example fix

# before
formatter.format([{"role": "user", "content": "hi"}])

# after
formatter.format([Msg.from_dict({"role": "user", "content": "hi"})])  # or construct Msg("user", "hi") directly

Defensive patterns

Strategy: validation

Validate before calling

msgs = [Msg.from_dict(m) if isinstance(m, dict) else m for m in msgs]
assert all(isinstance(m, Msg) for m in msgs)

Type guard

def is_msg_list(msgs) -> bool:
    return isinstance(msgs, list) and all(isinstance(m, Msg) for m in msgs)

Try / catch

try:
    formatter.format(msgs)
except TypeError as e:
    if "Expected Msg object" in str(e):
        msgs = [Msg.from_dict(m) if isinstance(m, dict) else m for m in msgs]
        formatted = formatter.format(msgs)
    else:
        raise

Prevention

Deserialize persisted conversations back to Msg before formatting
Type-annotate helper functions as list[Msg] so mypy catches this early

When it happens

Trigger: Passing a list like [msg, {"role": "user", "content": "hi"}] or [msg, "hi"]; loading a conversation from JSON and formatting it without deserializing dicts back to Msg objects.

Common situations: Restoring conversations from persistence/JSON; mixing API responses with local Msg objects; helper functions that return content strings instead of Msgs.

Understand the failure class

Background: "Must be a positive integer", "Invalid value", "Unsupported": the invalid-argument-value error family, when a library rejects the value you pass — this error's family across 35 libraries.

Related errors

AI-assisted analysis of agentscope-ai/agentscope@e90f1c7592 (2026-08-28). Data as JSON: /api/errors/2621861df18878a4. Report an issue: GitHub.

Appendix: source

Thrown at src/agentscope/formatter/_formatter_base.py:65

    @abstractmethod
    async def format(self, *args: Any, **kwargs: Any) -> list[dict[str, Any]]:
        """Format the Msg objects to a list of dictionaries that satisfy the
        API requirements."""

    @staticmethod
    def assert_list_of_msgs(msgs: list[Msg]) -> None:
        """Assert that the input is a list of Msg objects.

        Args:
            msgs (`list[Msg]`):
                A list of Msg objects to be validated.
        """
        if not isinstance(msgs, list):
            raise TypeError("Input must be a list of Msg objects.")

        for msg in msgs:
            if not isinstance(msg, Msg):
                raise TypeError(
                    f"Expected Msg object, got {type(msg)} instead.",
                )

    @staticmethod
    def _convert_unsupported_data_block_to_string(block: DataBlock) -> str:
        """Convert an unsupported data block into its textual fallback.

        URL sources remain accessible through their URL. Base64 sources are
        persisted to a temporary file so the model can still reference the
        result when the target API cannot carry that media type directly.

        Args:
            block (`DataBlock`):
                The unsupported data block to convert.

        Returns:
            `str`:
                The textual fallback for the data block.

View on GitHub (pinned to e90f1c7592)