GoalEvaluation

Build extensions, choose hooks, and manage request-scoped state. See the guide.

class zett_agent.extensions.goal.GoalEvaluation(*, achieved, summary, remaining_work=<factory>, next_instruction)

Structured decision reported by the private evaluator Agent.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

参数:

• achieved (bool)

• summary (GoalSummary)

• remaining_work (RemainingWork)

• next_instruction (NextInstruction)

validate_decision()

Keep terminal and continuation decisions internally consistent.

返回类型:

GoalEvaluation

Fields

achieved (required)

Whether the current result fully satisfies the original goal

summary (required)

Concise evidence-based explanation of the decision

remaining_work (optional)

Concrete unmet requirements; empty only when achieved is true

next_instruction (required)

Actionable instruction for the parent Agent; required when achieved is false