langchain-ai/deepagents · error · TypeError

Unknown keys in GeneralPurposeSubagentProfile dict: {sorted(

Error message

Unknown keys in GeneralPurposeSubagentProfile dict: {sorted(unknown)}

What it means

GeneralPurposeSubagentProfile.from_dict only accepts the fixed key set _GENERAL_PURPOSE_SUBAGENT_KEYS (enabled, description, system_prompt). Any other key in the dict raises TypeError, protecting users from silently ignored typos.

Source

Thrown at libs/deepagents/deepagents/profiles/harness/harness_profiles.py:175

    @classmethod
    def from_dict(cls, data: Mapping[str, Any]) -> GeneralPurposeSubagentProfile:
        """Construct a sub-profile from a plain dict.

        Args:
            data: Mapping with any subset of `enabled`, `description`, and
                `system_prompt` keys.

        Returns:
            A new `GeneralPurposeSubagentProfile`.

        Raises:
            TypeError: If `data` contains unknown keys, or if any value has
                the wrong type.
        """
        unknown = set(data.keys()) - _GENERAL_PURPOSE_SUBAGENT_KEYS
        if unknown:
            msg = f"Unknown keys in GeneralPurposeSubagentProfile dict: {sorted(unknown)}"
            raise TypeError(msg)
        enabled = data.get("enabled")
        description = data.get("description")
        system_prompt = data.get("system_prompt")
        if enabled is not None and not isinstance(enabled, bool):
            msg = f"`enabled` must be bool or None, got {type(enabled).__name__}"
            raise TypeError(msg)
        if description is not None and not isinstance(description, str):
            msg = f"`description` must be str or None, got {type(description).__name__}"
            raise TypeError(msg)
        if system_prompt is not None and not isinstance(system_prompt, str):
            msg = f"`system_prompt` must be str or None, got {type(system_prompt).__name__}"
            raise TypeError(msg)
        return cls(enabled=enabled, description=description, system_prompt=system_prompt)


@dataclass(frozen=True)
class HarnessProfileConfig:
    """Declarative harness-profile config for YAML/JSON-backed profiles.

View on GitHub (pinned to a1af029e6e)

Solutions

1. Remove unknown keys; keep only enabled, description, system_prompt.
2. Fix key spelling to snake_case ('system_prompt' not 'systemPrompt').
3. Check the schema/version expected by the installed deepagents release.

Example fix

// before
from_dict({"systemPrompt": "...", "verbose": True})
// after
from_dict({"system_prompt": "..."})

ALLOWED = {"enabled", "description", "system_prompt"}
unknown = set(data.keys()) - ALLOWED
if unknown:
    raise ValueError(f"unexpected keys: {sorted(unknown)}")

def is_subagent_dict(v: object) -> bool:
    return isinstance(v, dict) and set(v) <= {"enabled", "description", "system_prompt"}

try:
    profile = GeneralPurposeSubagentProfile.from_dict(data)
except TypeError as e:
    logging.error("invalid subagent profile config: %s", e)
    profile = GeneralPurposeSubagentProfile()

When it happens

Trigger: Passing a dict with misspelled or extra keys, e.g. {'systemPrompt': ...}, {'prompt': ...}, or a whole unrelated config block, to from_dict via _coerce_general_purpose_subagent.

Common situations: YAML/JSON harness config with camelCase instead of snake_case; leftover keys from an older schema version; nesting a provider config under the subagent block by mistake.

Understand the failure class

Background: Schema validation failed / invalid input schema: payload rejected because its shape doesn't match the expected schema — this error's family across 28 libraries.

Related errors

• Profile key must be a non-empty string.
• Profile key {key!r} has leading or trailing whitespace; expe
• Profile key {key!r} has more than one ':'; expected 'provide
• Profile key {key!r} has an empty provider or model half; exp
• Profile key {key!r} has whitespace adjacent to ':'; expected

AI-assisted analysis of langchain-ai/deepagents@a1af029e6e (2026-08-29). Data as JSON: /api/errors/2a7ff87bce642aea. Report an issue: GitHub.