Implementation:Elevenlabs Elevenlabs python UnitTestToolCallParameterEval

Attribute	Value
Page Type	Implementation
Package	elevenlabs
Module	elevenlabs.types.unit_test_tool_call_parameter_eval
Type	Discriminated Union (4 variants)
Base Class	UncheckedBaseModel (per variant)
Source File	src/elevenlabs/types/unit_test_tool_call_parameter_eval.py
Auto-Generated	Yes (Fern API Definition)

Overview

Description

UnitTestToolCallParameterEval is a discriminated union type that defines how a tool call parameter should be evaluated during unit testing. It supports four evaluation strategies, each represented as a separate Pydantic model variant:

• Anything -- Accepts any value for the parameter (no validation).
• Exact -- Requires the parameter to match an exact expected value.
• Llm -- Uses an LLM to evaluate the parameter based on a description.
• Regex -- Validates the parameter against a regular expression pattern.

The union is discriminated on the type field, which determines which variant is active.

Usage

This union type is used within tool call evaluation configurations to specify how individual parameters of a tool call should be validated. Each variant represents a different evaluation strategy, providing flexibility ranging from permissive (anything) to strict (exact match) to pattern-based (regex) to semantic (LLM-based) evaluation.

Code Reference

Source Location

src/elevenlabs/types/unit_test_tool_call_parameter_eval.py

Class Signatures

class UnitTestToolCallParameterEval_Anything(UncheckedBaseModel):
    type: typing.Literal["anything"] = "anything"

class UnitTestToolCallParameterEval_Exact(UncheckedBaseModel):
    type: typing.Literal["exact"] = "exact"
    expected_value: str

class UnitTestToolCallParameterEval_Llm(UncheckedBaseModel):
    type: typing.Literal["llm"] = "llm"
    description: str

class UnitTestToolCallParameterEval_Regex(UncheckedBaseModel):
    type: typing.Literal["regex"] = "regex"
    pattern: str

UnitTestToolCallParameterEval = Annotated[
    Union[
        UnitTestToolCallParameterEval_Anything,
        UnitTestToolCallParameterEval_Exact,
        UnitTestToolCallParameterEval_Llm,
        UnitTestToolCallParameterEval_Regex,
    ],
    UnionMetadata(discriminant="type"),
]

Import Statement

from elevenlabs.types.unit_test_tool_call_parameter_eval import (
    UnitTestToolCallParameterEval,
    UnitTestToolCallParameterEval_Anything,
    UnitTestToolCallParameterEval_Exact,
    UnitTestToolCallParameterEval_Llm,
    UnitTestToolCallParameterEval_Regex,
)

I/O Contract

Variant: Anything

Field Name	Type	Required	Default	Description
type	Literal["anything"]	Yes	"anything"	Discriminant field identifying this as an anything-match evaluation

Variant: Exact

Field Name	Type	Required	Default	Description
type	Literal["exact"]	Yes	"exact"	Discriminant field identifying this as an exact-match evaluation
expected_value	str	Yes	N/A	The exact string value the parameter must match

Variant: Llm

Field Name	Type	Required	Default	Description
type	Literal["llm"]	Yes	"llm"	Discriminant field identifying this as an LLM-based evaluation
description	str	Yes	N/A	A description prompt for the LLM to evaluate the parameter value

Variant: Regex

Field Name	Type	Required	Default	Description
type	Literal["regex"]	Yes	"regex"	Discriminant field identifying this as a regex-pattern evaluation
pattern	str	Yes	N/A	The regular expression pattern the parameter value must match

Usage Examples

Using Exact Match Evaluation

from elevenlabs.types.unit_test_tool_call_parameter_eval import (
    UnitTestToolCallParameterEval_Exact,
)

eval_exact = UnitTestToolCallParameterEval_Exact(
    expected_value="2025-01-15",
)
print(f"Type: {eval_exact.type}")  # "exact"
print(f"Expected: {eval_exact.expected_value}")

Using Regex Pattern Evaluation

from elevenlabs.types.unit_test_tool_call_parameter_eval import (
    UnitTestToolCallParameterEval_Regex,
)

eval_regex = UnitTestToolCallParameterEval_Regex(
    pattern=r"^\d{4}-\d{2}-\d{2}$",
)
print(f"Type: {eval_regex.type}")  # "regex"
print(f"Pattern: {eval_regex.pattern}")

Using LLM-Based Evaluation

from elevenlabs.types.unit_test_tool_call_parameter_eval import (
    UnitTestToolCallParameterEval_Llm,
)

eval_llm = UnitTestToolCallParameterEval_Llm(
    description="The value should be a valid future date in ISO format",
)

Using Anything Evaluation

from elevenlabs.types.unit_test_tool_call_parameter_eval import (
    UnitTestToolCallParameterEval_Anything,
)

eval_any = UnitTestToolCallParameterEval_Anything()
print(f"Type: {eval_any.type}")  # "anything"

Related Pages

• Elevenlabs_Elevenlabs_python_UnitTestToolCallEvaluationModelInput
• Elevenlabs_Elevenlabs_python_GetUnitTestResponseModel
• Elevenlabs_Elevenlabs_python_UnitTestCommonModel
• Elevenlabs_Elevenlabs_python_UnitTestSummaryResponseModel