langchain-ai/deepagents · error · TypeError

cli_max_retries must be None or an int, got {self.cli_max_re

Error message

cli_max_retries must be None or an int, got {self.cli_max_retries!r}

What it means

`ModelResult.__post_init__` raising TypeError because `cli_max_retries` is a bool instead of None or an int. Same rationale as model_retries: `True` would silently act as a retry budget of 1.

Source

Thrown at libs/code/deepagents_code/config.py:5578

        it was the one budget field in dcode without the check -- which is the
        argument for a single validated budget type rather than a ninth copy of
        this predicate.

        Raises:
            TypeError: If `model_retries` or `cli_max_retries` is a `bool`.
            ValueError: If `model_retries` or `cli_max_retries` is negative.
        """
        if isinstance(self.model_retries, bool):
            msg = f"model_retries must be an int, got {self.model_retries!r}"
            raise TypeError(msg)
        if self.model_retries < 0:
            msg = f"model_retries must be >= 0, got {self.model_retries}"
            raise ValueError(msg)
        if isinstance(self.cli_max_retries, bool):
            msg = (
                f"cli_max_retries must be None or an int, got {self.cli_max_retries!r}"
            )
            raise TypeError(msg)
        if self.cli_max_retries is not None and self.cli_max_retries < 0:
            msg = f"cli_max_retries must be >= 0, got {self.cli_max_retries}"
            raise ValueError(msg)

    def apply_to_runtime_state(self) -> None:
        """Commit this result's metadata to global `runtime_state`."""
        state = _get_runtime_state()
        state.model_name = self.model_name
        state.model_provider = self.provider
        state.model_context_limit = self.context_limit
        state.model_unsupported_modalities = self.unsupported_modalities


def _apply_profile_overrides(
    model: BaseChatModel,
    overrides: dict[str, Any],
    model_name: str,
    *,

View on GitHub (pinned to a1af029e6e)

Solutions

1. Pass an int or None for `cli_max_retries`, never a bool.
2. Coerce config-sourced values with a validator that rejects bools before construction.
3. Use None to mean 'flag not set' rather than False.

Example fix

// before
ModelResult(..., cli_max_retries=False)
// after
ModelResult(..., cli_max_retries=None)  # or an int like 3

def is_cli_max_retries(v: object) -> bool:
    return v is None or (isinstance(v, int) and not isinstance(v, bool) and v >= 0)
assert is_cli_max_retries(cli_max_retries)

def is_optional_int(v: object) -> TypeGuard[int | None]:
    return v is None or (isinstance(v, int) and not isinstance(v, bool))

try:
    result = ModelResult(..., cli_max_retries=v)
except TypeError:
    result = ModelResult(..., cli_max_retries=None if isinstance(v, bool) else v)

When it happens

Trigger: Constructing `ModelResult(...)` with `cli_max_retries=True`/`False`. The CLI flag path enforces `non_negative_int`, so this fires only when the value bypasses argparse (programmatic construction, hand-wired config plumbing).

Common situations: Tests or tooling constructing ModelResult directly; a bool leaking from a config parser that accepts true/false.

Related errors

• model_retries must be an int, got {self.model_retries!r}
• interpreter_ptc must be False, 'safe', 'all', or a list of t
• model_retries must be >= 0, got {self.model_retries}
• cli_max_retries must be >= 0, got {self.cli_max_retries}
• Goal criteria request must be an object.

AI-assisted analysis of langchain-ai/deepagents@a1af029e6e (2026-08-29). Data as JSON: /api/errors/b07961832ac11109. Report an issue: GitHub.