langchain-ai/deepagents · error · ValueError

goal application requires a non-empty objective and rubric

Error message

goal application requires a non-empty objective and rubric

What it means

A goal application (objective + rubric pair) was constructed with an empty `objective` or `rubric`. The dataclass `__post_init__` enforces that both fields are non-empty before delegating to `validate_goal_application` for size limits, failing fast on structurally useless goal definitions.

Source

Thrown at libs/code/deepagents_code/app.py:2349

    def __post_init__(self) -> None:
        """Reject an empty or oversized objective or rubric at construction.

        Callers already screen these (`_accept_goal_rubric`), so this guards
        against a future construction site skipping that check — queuing a goal
        that would clear the active one to nothing, or one whose text cannot
        render as a goal-state notice. Holding both invariants here means they
        do not depend on every construction site remembering; callers keep their
        own checks so they can report a targeted message before mutating UI
        state.

        Raises:
            ValueError: If `objective` or `rubric` is empty.
            GoalStateSizeError: If either exceeds its own limit, or their
                combined text exceeds `GOAL_APPLICATION_CHAR_LIMIT`.
        """  # noqa: DOC502 - propagates from `validate_goal_application`
        if not self.objective or not self.rubric:
            msg = "goal application requires a non-empty objective and rubric"
            raise ValueError(msg)
        validate_goal_application(self.objective, self.rubric)


@dataclass(frozen=True, slots=True)
class _GoalApplicationResult:
    """Applied goal transition and whether its checkpoint write succeeded."""

    transition: Literal["create", "amended"]
    persisted: bool


@dataclass(frozen=True, slots=True)
class _BlockedGoalResetResult:
    """Blocked-goal reset outcome used to gate the following user turn.

    Build instances through `proceed`, `reset`, or `failed` so only valid
    field combinations are constructed. `did_reset` — not a `blocker_note`
    sentinel — is the authoritative signal that a blocked goal was flipped

View on GitHub (pinned to a1af029e6e)

Solutions

1. Supply a non-empty objective and a non-empty rubric before constructing the object
2. Strip whitespace and reject empty strings at the input boundary
3. If a goal is genuinely absent, skip construction instead of passing empty strings

Example fix

// before
GoalApplication(objective="", rubric="answer correctly")
// after
if not objective.strip() or not rubric.strip():
    raise ValueError("objective and rubric are required")
GoalApplication(objective=objective, rubric=rubric)

if not objective or not objective.strip() or not rubric or not rubric.strip():
    raise ValueError("goal application requires a non-empty objective and rubric")

def has_goal_fields(goal: object) -> bool:
    return (
        isinstance(goal, GoalApplication)
        and bool(goal.objective)
        and bool(goal.rubric)
    )

try:
    goal = GoalApplication(objective=obj, rubric=rub)
except ValueError as exc:
    logger.warning("invalid goal: %s", exc)
    return None

When it happens

Trigger: Constructing the goal-application dataclass with `objective=''`, `rubric=''`, or None — typically from empty form fields, a config file with blank keys, or a programmatic call that passed unvalidated user input.

Common situations: Automation that builds goals from config/templates where a placeholder was never filled; user submitted a goal form with one field blank.

Understand the failure class

Background: "Missing required field" and "field is required" errors: why libraries reject payloads that omit mandatory fields — this error's family across 20 libraries.

Related errors

• shell_allow_list must be None or non-empty
• RewarmEstimate costs must be finite, got cold={self.cold_cos
• RewarmEstimate costs must be non-negative, got cold={self.co
• RewarmEstimate incremental cost {self.incremental_cost_usd!r
• ColdCacheWarning age_seconds={self.age_seconds!r} does not p

AI-assisted analysis of langchain-ai/deepagents@a1af029e6e (2026-08-29). Data as JSON: /api/errors/0f272f33b02bf668. Report an issue: GitHub.