ValidationError in structured_llm.invoke when generating analysts: Missing required fields (affiliation and description)

[Mu7annad0]

I am encountering a ValidationError when using the structured_llm.invoke method to generate a list of analysts. The error indicates that the required fields affiliation and description are missing for each analyst in the Perspective model. Here's the relevant code snippet:

analysts_response = structured_llm.invoke(
    [SystemMessage(content=system_message)] + [HumanMessage(content=human_msg)])

Error Message:

ValidationError: 6 validation errors for Perspective
analysts.0.affiliation
Field required [type=missing, input_value={'name': 'Dr. Anya Sharma...e': 'Lead AI Scientist'}, input_type=dict]
For further information visit https://errors.pydantic.dev/2.10/v/missing
analysts.0.description
Field required [type=missing, input_value={'name': 'Dr. Anya Sharma...e': 'Lead AI Scientist'}, input_type=dict]
For further information visit https://errors.pydantic.dev/2.10/v/missing
analysts.1.affiliation
Field required [type=missing, input_value={'role': 'Chief Software ...name': 'Mr. Ben Carter'}, input_type=dict]
For further information visit https://errors.pydantic.dev/2.10/v/missing
analysts.1.description
Field required [type=missing, input_value={'role': 'Chief Software ...name': 'Mr. Ben Carter'}, input_type=dict]
For further information visit https://errors.pydantic.dev/2.10/v/missing
analysts.2.affiliation
Field required [type=missing, input_value={'role': 'Senior Tech Pol...ame': 'Ms. Chloe Davis'}, input_type=dict]
For further information visit https://errors.pydantic.dev/2.10/v/missing
analysts.2.description
Field required [type=missing, input_value={'role': 'Senior Tech Pol...ame': 'Ms. Chloe Davis'}, input_type=dict]
For further information visit https://errors.pydantic.dev/2.10/v/missing

Expected Behavior:
The structured_llm.invoke method should return a valid Perspective object with all required fields (affiliation, name, role, and description) populated for each analyst.

Actual Behavior:
The method fails with a ValidationError because the affiliation and description fields are missing in the generated output.

Code Snippet:

class Analyst(BaseModel):
    affiliation: str = Field(
        description="Primary affiliation of the analyst.",
    )
    name: str = Field(
        description="Name of the analyst."
    )
    role: str = Field(
        description="Role of the analyst in the context of the topic.",
    )
    description: str = Field(
        description="Description of the analyst focus, concerns, and motives.",
    )
    @property
    def persona(self) -> str:
        return f"Name: {self.name}\nRole: {self.role}\nAffiliation: {self.affiliation}\nDescription: {self.description}\n"

class Perspective(BaseModel):
    analysts: List[Analyst] = Field(
        description="Comprehensive list of analysts with their roles and affiliations.",
    )

def create_analysts(state: GenerateAnalystsState):
    """Create analysts for a given topic"""

    topic = state["topic"]
    max_analysts = state["max_analysts"]
    human_analyst_feedback = state.get("human_analyst_feedback", '')

    structured_llm = llm.with_structured_output(Perspective)

    system_message = analyst_instructions.format(
        topic=topic,
        human_analyst_feedback=human_analyst_feedback,
        max_analysts=max_analysts,
    )

    human_msg = """Generate the set of analysts."""

    analysts_response = structured_llm.invoke(
        [SystemMessage(content=system_message)] + [HumanMessage(content=human_msg)]
    )

    return {
        "analysts": analysts_response.analysts
    }