scale-gp-beta 0.1.0a32__tar.gz → 0.1.0a33__tar.gz

scale_gp_beta-0.1.0a33/.release-please-manifest.json

@@ -0,0 +1,3 @@
+{
+  ".": "0.1.0-alpha.33"
+}

{scale_gp_beta-0.1.0a32 → scale_gp_beta-0.1.0a33}/CHANGELOG.md

@@ -1,5 +1,15 @@
 # Changelog
 
+## 0.1.0-alpha.33 (2025-10-03)
+
+Full Changelog: [v0.1.0-alpha.32...v0.1.0-alpha.33](https://github.com/scaleapi/sgp-python-beta/compare/v0.1.0-alpha.32...v0.1.0-alpha.33)
+
+### Features
+
+* **api:** api update ([78cd9ff](https://github.com/scaleapi/sgp-python-beta/commit/78cd9fff9644ccdb90fd31fc944da5eba11c9044))
+* **api:** api update ([e47c322](https://github.com/scaleapi/sgp-python-beta/commit/e47c3224f983ed125379b0bf98f947a0d8520fa1))
+* **api:** api update ([f0a9e63](https://github.com/scaleapi/sgp-python-beta/commit/f0a9e63cd22ecb8c4940cbffac94b97837fe762b))
+
 ## 0.1.0-alpha.32 (2025-09-26)
 
 Full Changelog: [v0.1.0-alpha.31...v0.1.0-alpha.32](https://github.com/scaleapi/sgp-python-beta/compare/v0.1.0-alpha.31...v0.1.0-alpha.32)

{scale_gp_beta-0.1.0a32 → scale_gp_beta-0.1.0a33}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: scale-gp-beta
-Version: 0.1.0a32
+Version: 0.1.0a33
 Summary: The official Python library for the Scale GP API
 Project-URL: Homepage, https://github.com/scaleapi/sgp-python-beta
 Project-URL: Repository, https://github.com/scaleapi/sgp-python-beta

{scale_gp_beta-0.1.0a32 → scale_gp_beta-0.1.0a33}/api.md

@@ -66,7 +66,7 @@ Methods:
 Types:
 
 ```python
-from scale_gp_beta.types import Question, QuestionList
+from scale_gp_beta.types import Question
 ```
 
 Methods:
@@ -80,7 +80,7 @@ Methods:
 Types:
 
 ```python
-from scale_gp_beta.types import File, FileList, FileDeleteResponse
+from scale_gp_beta.types import File, FileDeleteResponse
 ```
 
 Methods:
@@ -102,7 +102,7 @@ Methods:
 Types:
 
 ```python
-from scale_gp_beta.types import InferenceModel, InferenceModelList, ModelDeleteResponse
+from scale_gp_beta.types import InferenceModel, ModelDeleteResponse
 ```
 
 Methods:

{scale_gp_beta-0.1.0a32 → scale_gp_beta-0.1.0a33}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "scale-gp-beta"
-version = "0.1.0-alpha.32"
+version = "0.1.0-alpha.33"
 description = "The official Python library for the Scale GP API"
 dynamic = ["readme"]
 license = "Apache-2.0"

{scale_gp_beta-0.1.0a32 → scale_gp_beta-0.1.0a33}/src/scale_gp_beta/_version.py

@@ -1,4 +1,4 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 __title__ = "scale_gp_beta"
-__version__ = "0.1.0-alpha.32"  # x-release-please-version
+__version__ = "0.1.0-alpha.33"  # x-release-please-version

{scale_gp_beta-0.1.0a32 → scale_gp_beta-0.1.0a33}/src/scale_gp_beta/resources/evaluations.py

@@ -115,7 +115,7 @@ class EvaluationsResource(SyncAPIResource):
         Args:
           dataset_id: The ID of the dataset containing the items referenced by the `data` field
 
-          data: Items to be evaluated, including references to the input dataset items
+          data: Items to be evaluated, including references to the input dataset
 
           metadata: Optional metadata key-value pairs for the evaluation
 
@@ -480,7 +480,7 @@ class AsyncEvaluationsResource(AsyncAPIResource):
         Args:
           dataset_id: The ID of the dataset containing the items referenced by the `data` field
 
-          data: Items to be evaluated, including references to the input dataset items
+          data: Items to be evaluated, including references to the input dataset
 
           metadata: Optional metadata key-value pairs for the evaluation
 

{scale_gp_beta-0.1.0a32 → scale_gp_beta-0.1.0a33}/src/scale_gp_beta/resources/spans.py

@@ -256,10 +256,13 @@ class SpansResource(SyncAPIResource):
         starting_after: str | Omit = omit,
         to_ts: Union[str, datetime] | Omit = omit,
         application_variant_ids: SequenceNotStr[str] | Omit = omit,
+        assessment_types: SequenceNotStr[str] | Omit = omit,
         excluded_span_ids: SequenceNotStr[str] | Omit = omit,
         excluded_trace_ids: SequenceNotStr[str] | Omit = omit,
         extra_metadata: Dict[str, object] | Omit = omit,
         group_id: str | Omit = omit,
+        max_duration_ms: int | Omit = omit,
+        min_duration_ms: int | Omit = omit,
         names: SequenceNotStr[str] | Omit = omit,
         parents_only: bool | Omit = omit,
         search_texts: SequenceNotStr[str] | Omit = omit,
@@ -284,6 +287,8 @@ class SpansResource(SyncAPIResource):
 
           application_variant_ids: Filter by application variant IDs
 
+          assessment_types: Filter spans by traces that have assessments of these types
+
           excluded_span_ids: List of span IDs to exclude from results
 
           excluded_trace_ids: List of trace IDs to exclude from results
@@ -292,6 +297,10 @@ class SpansResource(SyncAPIResource):
 
           group_id: Filter by group ID
 
+          max_duration_ms: Maximum span duration in milliseconds (inclusive)
+
+          min_duration_ms: Minimum span duration in milliseconds (inclusive)
+
           names: Filter by trace/span name
 
           parents_only: Only fetch spans that are the top-level (ie. have no parent_id)
@@ -318,10 +327,13 @@ class SpansResource(SyncAPIResource):
             body=maybe_transform(
                 {
                     "application_variant_ids": application_variant_ids,
+                    "assessment_types": assessment_types,
                     "excluded_span_ids": excluded_span_ids,
                     "excluded_trace_ids": excluded_trace_ids,
                     "extra_metadata": extra_metadata,
                     "group_id": group_id,
+                    "max_duration_ms": max_duration_ms,
+                    "min_duration_ms": min_duration_ms,
                     "names": names,
                     "parents_only": parents_only,
                     "search_texts": search_texts,
@@ -604,10 +616,13 @@ class AsyncSpansResource(AsyncAPIResource):
         starting_after: str | Omit = omit,
         to_ts: Union[str, datetime] | Omit = omit,
         application_variant_ids: SequenceNotStr[str] | Omit = omit,
+        assessment_types: SequenceNotStr[str] | Omit = omit,
         excluded_span_ids: SequenceNotStr[str] | Omit = omit,
         excluded_trace_ids: SequenceNotStr[str] | Omit = omit,
         extra_metadata: Dict[str, object] | Omit = omit,
         group_id: str | Omit = omit,
+        max_duration_ms: int | Omit = omit,
+        min_duration_ms: int | Omit = omit,
         names: SequenceNotStr[str] | Omit = omit,
         parents_only: bool | Omit = omit,
         search_texts: SequenceNotStr[str] | Omit = omit,
@@ -632,6 +647,8 @@ class AsyncSpansResource(AsyncAPIResource):
 
           application_variant_ids: Filter by application variant IDs
 
+          assessment_types: Filter spans by traces that have assessments of these types
+
           excluded_span_ids: List of span IDs to exclude from results
 
           excluded_trace_ids: List of trace IDs to exclude from results
@@ -640,6 +657,10 @@ class AsyncSpansResource(AsyncAPIResource):
 
           group_id: Filter by group ID
 
+          max_duration_ms: Maximum span duration in milliseconds (inclusive)
+
+          min_duration_ms: Minimum span duration in milliseconds (inclusive)
+
           names: Filter by trace/span name
 
           parents_only: Only fetch spans that are the top-level (ie. have no parent_id)
@@ -666,10 +687,13 @@ class AsyncSpansResource(AsyncAPIResource):
             body=maybe_transform(
                 {
                     "application_variant_ids": application_variant_ids,
+                    "assessment_types": assessment_types,
                     "excluded_span_ids": excluded_span_ids,
                     "excluded_trace_ids": excluded_trace_ids,
                     "extra_metadata": extra_metadata,
                     "group_id": group_id,
+                    "max_duration_ms": max_duration_ms,
+                    "min_duration_ms": min_duration_ms,
                     "names": names,
                     "parents_only": parents_only,
                     "search_texts": search_texts,

{scale_gp_beta-0.1.0a32 → scale_gp_beta-0.1.0a33}/src/scale_gp_beta/types/__init__.py

@@ -12,14 +12,12 @@ from .question import Question as Question
 from .response import Response as Response
 from .component import Component as Component
 from .container import Container as Container
-from .file_list import FileList as FileList
 from .span_type import SpanType as SpanType
 from .completion import Completion as Completion
 from .evaluation import Evaluation as Evaluation
 from .span_status import SpanStatus as SpanStatus
 from .dataset_item import DatasetItem as DatasetItem
 from .item_locator import ItemLocator as ItemLocator
-from .question_list import QuestionList as QuestionList
 from .component_param import ComponentParam as ComponentParam
 from .container_param import ContainerParam as ContainerParam
 from .evaluation_item import EvaluationItem as EvaluationItem
@@ -39,7 +37,6 @@ from .model_create_params import ModelCreateParams as ModelCreateParams
 from .model_update_params import ModelUpdateParams as ModelUpdateParams
 from .span_batch_response import SpanBatchResponse as SpanBatchResponse
 from .file_delete_response import FileDeleteResponse as FileDeleteResponse
-from .inference_model_list import InferenceModelList as InferenceModelList
 from .question_list_params import QuestionListParams as QuestionListParams
 from .dataset_create_params import DatasetCreateParams as DatasetCreateParams
 from .dataset_update_params import DatasetUpdateParams as DatasetUpdateParams

{scale_gp_beta-0.1.0a32 → scale_gp_beta-0.1.0a33}/src/scale_gp_beta/types/dataset_delete_response.py

@@ -1,8 +1,5 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-from typing import Optional
-from typing_extensions import Literal
-
 from .._models import BaseModel
 
 __all__ = ["DatasetDeleteResponse"]
@@ -13,4 +10,4 @@ class DatasetDeleteResponse(BaseModel):
 
     deleted: bool
 
-    object: Optional[Literal["dataset"]] = None
+    object: str

{scale_gp_beta-0.1.0a32 → scale_gp_beta-0.1.0a33}/src/scale_gp_beta/types/dataset_item_delete_response.py

@@ -1,8 +1,5 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-from typing import Optional
-from typing_extensions import Literal
-
 from .._models import BaseModel
 
 __all__ = ["DatasetItemDeleteResponse"]
@@ -13,4 +10,4 @@ class DatasetItemDeleteResponse(BaseModel):
 
     deleted: bool
 
-    object: Optional[Literal["dataset.item"]] = None
+    object: str

{scale_gp_beta-0.1.0a32 → scale_gp_beta-0.1.0a33}/src/scale_gp_beta/types/evaluation_create_params.py

@@ -42,7 +42,7 @@ class EvaluationFromDatasetCreateRequest(TypedDict, total=False):
     name: Required[str]
 
     data: Iterable[EvaluationFromDatasetCreateRequestData]
-    """Items to be evaluated, including references to the input dataset items"""
+    """Items to be evaluated, including references to the input dataset"""
 
     description: str
 

{scale_gp_beta-0.1.0a32 → scale_gp_beta-0.1.0a33}/src/scale_gp_beta/types/evaluation_delete_response.py

@@ -1,8 +1,5 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-from typing import Optional
-from typing_extensions import Literal
-
 from .._models import BaseModel
 
 __all__ = ["EvaluationDeleteResponse"]
@@ -13,4 +10,4 @@ class EvaluationDeleteResponse(BaseModel):
 
     deleted: bool
 
-    object: Optional[Literal["evaluation"]] = None
+    object: str

{scale_gp_beta-0.1.0a32 → scale_gp_beta-0.1.0a33}/src/scale_gp_beta/types/evaluation_task.py

@@ -41,6 +41,27 @@ __all__ = [
     "AutoEvaluationGuidedDecodingEvaluationTaskConfiguration",
     "AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationStructuredOutputTaskRequestWithItemLocator",
     "AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationGuidedDecodingTaskRequestWithItemLocator",
+    "AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocator",
+    "AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedTo",
+    "AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToApeAgent",
+    "AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToApeAgentConfig",
+    "AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToIfAgent",
+    "AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToIfAgentConfig",
+    "AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToTruthfulnessAgent",
+    "AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToTruthfulnessAgentConfig",
+    "AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToBaseAgent",
+    "AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToBaseAgentConfig",
+    "AutoEvaluationAgentEvaluationTask",
+    "AutoEvaluationAgentEvaluationTaskConfiguration",
+    "AutoEvaluationAgentEvaluationTaskConfigurationDesignatedTo",
+    "AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToApeAgent",
+    "AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToApeAgentConfig",
+    "AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToIfAgent",
+    "AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToIfAgentConfig",
+    "AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToTruthfulnessAgent",
+    "AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToTruthfulnessAgentConfig",
+    "AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToBaseAgent",
+    "AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToBaseAgentConfig",
     "ContributorEvaluationQuestionTask",
     "ContributorEvaluationQuestionTaskConfiguration",
 ]
@@ -368,9 +389,94 @@ class AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationGuide
     system_prompt: Optional[str] = None
 
 
+class AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToApeAgentConfig(
+    BaseModel
+):
+    model: Optional[str] = None
+
+    temperature: Optional[float] = None
+
+
+class AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToApeAgent(
+    BaseModel
+):
+    config: AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToApeAgentConfig
+
+    agent_name: Optional[Literal["APEAgent"]] = None
+
+
+class AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToIfAgentConfig(
+    BaseModel
+):
+    model: Optional[str] = None
+
+
+class AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToIfAgent(
+    BaseModel
+):
+    config: AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToIfAgentConfig
+
+    agent_name: Optional[Literal["IFAgent"]] = None
+
+
+class AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToTruthfulnessAgentConfig(
+    BaseModel
+):
+    model: Optional[str] = None
+
+
+class AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToTruthfulnessAgent(
+    BaseModel
+):
+    config: AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToTruthfulnessAgentConfig
+
+    agent_name: Optional[Literal["TruthfulnessAgent"]] = None
+
+
+class AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToBaseAgentConfig(
+    BaseModel
+):
+    model: Optional[str] = None
+
+
+class AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToBaseAgent(
+    BaseModel
+):
+    config: AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToBaseAgentConfig
+
+    agent_name: Optional[Literal["BaseAgent"]] = None
+
+
+AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedTo: TypeAlias = Union[
+    AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToApeAgent,
+    AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToIfAgent,
+    AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToTruthfulnessAgent,
+    AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedToBaseAgent,
+]
+
+
+class AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocator(BaseModel):
+    definition: str
+
+    name: str
+
+    output_rules: List[str]
+
+    data_fields: Optional[List[str]] = None
+
+    designated_to: Optional[
+        AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocatorDesignatedTo
+    ] = None
+
+    output_type: Optional[Literal["text", "integer", "float", "boolean"]] = None
+
+    output_values: Optional[List[Union[str, float, bool]]] = None
+
+
 AutoEvaluationGuidedDecodingEvaluationTaskConfiguration: TypeAlias = Union[
     AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationStructuredOutputTaskRequestWithItemLocator,
     AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationGuidedDecodingTaskRequestWithItemLocator,
+    AutoEvaluationGuidedDecodingEvaluationTaskConfigurationAutoEvaluationAgentTaskRequestWithItemLocator,
 ]
 
 
@@ -386,6 +492,81 @@ class AutoEvaluationGuidedDecodingEvaluationTask(BaseModel):
     task_type: Optional[Literal["auto_evaluation.guided_decoding"]] = None
 
 
+class AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToApeAgentConfig(BaseModel):
+    model: Optional[str] = None
+
+    temperature: Optional[float] = None
+
+
+class AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToApeAgent(BaseModel):
+    config: AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToApeAgentConfig
+
+    agent_name: Optional[Literal["APEAgent"]] = None
+
+
+class AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToIfAgentConfig(BaseModel):
+    model: Optional[str] = None
+
+
+class AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToIfAgent(BaseModel):
+    config: AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToIfAgentConfig
+
+    agent_name: Optional[Literal["IFAgent"]] = None
+
+
+class AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToTruthfulnessAgentConfig(BaseModel):
+    model: Optional[str] = None
+
+
+class AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToTruthfulnessAgent(BaseModel):
+    config: AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToTruthfulnessAgentConfig
+
+    agent_name: Optional[Literal["TruthfulnessAgent"]] = None
+
+
+class AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToBaseAgentConfig(BaseModel):
+    model: Optional[str] = None
+
+
+class AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToBaseAgent(BaseModel):
+    config: AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToBaseAgentConfig
+
+    agent_name: Optional[Literal["BaseAgent"]] = None
+
+
+AutoEvaluationAgentEvaluationTaskConfigurationDesignatedTo: TypeAlias = Union[
+    AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToApeAgent,
+    AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToIfAgent,
+    AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToTruthfulnessAgent,
+    AutoEvaluationAgentEvaluationTaskConfigurationDesignatedToBaseAgent,
+]
+
+
+class AutoEvaluationAgentEvaluationTaskConfiguration(BaseModel):
+    definition: str
+
+    name: str
+
+    output_rules: List[str]
+
+    data_fields: Optional[List[str]] = None
+
+    designated_to: Optional[AutoEvaluationAgentEvaluationTaskConfigurationDesignatedTo] = None
+
+    output_type: Optional[Literal["text", "integer", "float", "boolean"]] = None
+
+    output_values: Optional[List[Union[str, float, bool]]] = None
+
+
+class AutoEvaluationAgentEvaluationTask(BaseModel):
+    configuration: AutoEvaluationAgentEvaluationTaskConfiguration
+
+    alias: Optional[str] = None
+    """Alias to title the results column. Defaults to the `auto_evaluation_agent`"""
+
+    task_type: Optional[Literal["auto_evaluation.agent"]] = None
+
+
 class ContributorEvaluationQuestionTaskConfiguration(BaseModel):
     layout: "Container"
 
@@ -415,6 +596,7 @@ EvaluationTask: TypeAlias = Annotated[
         MetricEvaluationTask,
         AutoEvaluationQuestionTask,
         AutoEvaluationGuidedDecodingEvaluationTask,
+        AutoEvaluationAgentEvaluationTask,
         ContributorEvaluationQuestionTask,
     ],
     PropertyInfo(discriminator="task_type"),