> ## Documentation Index > Fetch the complete documentation index at: https://www.truefoundry.com/llms.txt > Use this file to discover all available pages before exploring further. # Get artifact > Get an artifact by its ID. ## OpenAPI ````yaml /openapi.json get /api/ml/v1/artifacts/{id} openapi: 3.1.0 info: title: TrueFoundry API description: API for TrueFoundry version: 0.1.0 contact: {} servers: - url: https://{controlPlaneURL} variables: controlPlaneURL: default: app.truefoundry.com description: Control Plane URL security: [] tags: - name: Agent - name: Agent Skills - name: Applications - name: Metrics - name: Apply - name: Artifacts - name: Audit Logs - name: Clusters - name: Jobs - name: Logs - name: MCP Servers V2 - name: MLRepos - name: Model Deployments - name: Models - name: Personal Access Tokens - name: Prompts - name: Provider Integrations - name: SCIM v2 - name: Secret Groups - name: Secrets - name: Teams - name: Traces - name: Users - name: Virtual Accounts - name: Role Bindings - name: Workspaces paths: /api/ml/v1/artifacts/{id}: get: tags: - Artifacts summary: Get artifact description: Get an artifact by its ID. operationId: get_artifact_v1_artifacts__id__get parameters: - required: true schema: type: string title: Id name: id in: path responses: '200': description: The artifact data content: application/json: schema: $ref: '#/components/schemas/GetArtifactResponse' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - jwt: [] components: schemas: GetArtifactResponse: properties: data: allOf: - $ref: '#/components/schemas/Artifact' title: Data description: The artifact data type: object required: - data title: GetArtifactResponse x-fern-type-name: GetArtifactResponse HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError x-fern-type-name: HTTPValidationError Artifact: properties: id: type: string title: Id description: Unique identifier for the artifact ml_repo_id: type: string title: Ml Repo Id description: ID of the ML Repo that this artifact belongs to type: allOf: - $ref: '#/components/schemas/ArtifactType' description: >- Type of the artifact (e.g., 'artifact', 'model', 'chat_prompt', 'agent-skill', 'plot', 'image') name: type: string title: Name description: >- Name of the artifact (alphanumeric characters, hyphens, and underscores only, max 256 characters) fqn: type: string title: Fqn description: >- Fully qualified name of the artifact in the format '{artifact_type}:{tenant_name}/{ml_repo_name}/{artifact_name}' created_by_subject: allOf: - $ref: '#/components/schemas/Subject' title: Created By Subject description: Subject (user, team, or service account) that created this artifact created_at: type: string format: date-time title: Created At description: Timestamp when the artifact was created updated_at: type: string format: date-time title: Updated At description: Timestamp when the artifact was last updated latest_version: allOf: - $ref: '#/components/schemas/BaseArtifactVersion' title: Latest Version description: The most recent version of this artifact run_steps: items: type: integer type: array title: Run Steps description: List of run step numbers where this artifact was created or updated type: object required: - id - ml_repo_id - type - name - fqn - created_by_subject title: Artifact description: Generic artifact API DTO including run-step linkage for ML runs. x-fern-type-name: Artifact ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type type: object required: - loc - msg - type title: ValidationError x-fern-type-name: ValidationError ArtifactType: type: string enum: - artifact - model - plot - image - chat_prompt - agent-skill title: ArtifactType description: An enumeration. x-fern-type-name: ArtifactType Subject: type: object properties: subjectId: type: string description: Subject ID subjectType: description: Subject type allOf: - $ref: '#/components/schemas/SubjectType' subjectSlug: type: string description: Subject slug nullable: true subjectDisplayName: type: string description: Subject display name nullable: true subjectPatName: type: string description: Subject PAT name nullable: true subjectControllerName: type: string nullable: true subjectExternalIdentitySlug: type: string description: External identity slug (external_identity_id:sub:email) nullable: true required: - subjectId - subjectType title: Subject x-fern-type-name: Subject BaseArtifactVersion: properties: id: type: string title: Id description: Unique identifier for the artifact version fqn: type: string title: Fqn description: >- Fully qualified name of the artifact version in the format '{artifact_type}:{tenant_name}/{ml_repo_name}/{artifact_name}:{version}' created_by_subject: allOf: - $ref: '#/components/schemas/Subject' title: Created By Subject description: >- Subject (user, team, or service account) that created this artifact version created_at: type: string format: date-time title: Created At description: Timestamp when the artifact version was created updated_at: type: string format: date-time title: Updated At description: Timestamp when the artifact version was last updated manifest: anyOf: - $ref: '#/components/schemas/ModelManifest' - $ref: '#/components/schemas/ChatPromptManifest' - $ref: '#/components/schemas/ArtifactManifest' - $ref: '#/components/schemas/AgentSkillManifest' title: Manifest ml_repo_id: type: string title: Ml Repo Id description: ID of the ML Repo that this artifact version belongs to type: object required: - id - fqn - created_by_subject - ml_repo_id title: BaseArtifactVersion description: Shared artifact-version fields (identity, manifest union, ML repo). x-fern-type-name: BaseArtifactVersion SubjectType: type: string enum: - user - team - serviceaccount - virtualaccount - external-identity - agent-identity - role description: >- Type of the authenticated subject. "user" for human users, "serviceaccount" (also known as virtual account) for programmatic access. title: SubjectType x-fern-type-name: SubjectType ModelManifest: properties: name: type: string maxLength: 256 pattern: ^[A-Za-z0-9_\-]+$ title: Name description: >- Name of the model (alphanumeric characters, hyphens, and underscores only, max 256 characters) metadata: type: object title: Metadata description: >- Key value metadata. Should be valid JSON. For e.g. `{"business-unit": "sales", "quality": "good", "rating": 4.5}` ml_repo: type: string pattern: ^[a-zA-Z][a-zA-Z0-9\-]{1,98}[a-zA-Z0-9]$ title: Ml Repo description: >- Name of the ML Repo that this model belongs to (must start and end with alphanumeric, 2-100 characters) version: type: integer minimum: 1 title: Version description: Version of the entity type: type: string enum: - model-version title: Type default: model-version description: type: string maxLength: 512 title: Description version_alias: type: string maxLength: 128 pattern: ^v[a-zA-Z0-9.-]*([a-zA-Z0-9]+)$ title: Version Alias description: >- Version alias is alternate, ideally human readable, version string to reference an artifact version. It should start with `v` followed by alphanumeric and it can include `.` and `-` in between (e.g. `v1.0.0`, `v1-prod`, `v3-dev`, etc) source: oneOf: - $ref: '#/components/schemas/TrueFoundryManagedSource' - $ref: '#/components/schemas/ExternalBlobStorageSource' - $ref: '#/components/schemas/LocalModelSource' title: Source discriminator: propertyName: type mapping: truefoundry: $ref: '#/components/schemas/TrueFoundryManagedSource' external: $ref: '#/components/schemas/ExternalBlobStorageSource' local: $ref: '#/components/schemas/LocalModelSource' framework: oneOf: - $ref: '#/components/schemas/TransformersFramework' - $ref: '#/components/schemas/TensorFlowFramework' - $ref: '#/components/schemas/SklearnFramework' - $ref: '#/components/schemas/PyTorchFramework' - $ref: '#/components/schemas/KerasFramework' - $ref: '#/components/schemas/XGBoostFramework' - $ref: '#/components/schemas/LightGBMFramework' - $ref: '#/components/schemas/FastAIFramework' - $ref: '#/components/schemas/H2OFramework' - $ref: '#/components/schemas/ONNXFramework' - $ref: '#/components/schemas/SpaCyFramework' - $ref: '#/components/schemas/StatsModelsFramework' - $ref: '#/components/schemas/GluonFramework' - $ref: '#/components/schemas/PaddleFramework' title: Framework description: >- Framework for the model version like Transformers, PyTorch, Sklearn, Xgboost etc with framework specific metadata. This will be used to infer model deployment configuration discriminator: propertyName: type mapping: transformers: $ref: '#/components/schemas/TransformersFramework' tensorflow: $ref: '#/components/schemas/TensorFlowFramework' sklearn: $ref: '#/components/schemas/SklearnFramework' pytorch: $ref: '#/components/schemas/PyTorchFramework' keras: $ref: '#/components/schemas/KerasFramework' xgboost: $ref: '#/components/schemas/XGBoostFramework' lightgbm: $ref: '#/components/schemas/LightGBMFramework' fastai: $ref: '#/components/schemas/FastAIFramework' h2o: $ref: '#/components/schemas/H2OFramework' onnx: $ref: '#/components/schemas/ONNXFramework' spacy: $ref: '#/components/schemas/SpaCyFramework' statsmodels: $ref: '#/components/schemas/StatsModelsFramework' gluon: $ref: '#/components/schemas/GluonFramework' paddle: $ref: '#/components/schemas/PaddleFramework' environment: $ref: '#/components/schemas/ModelVersionEnvironment' step: type: integer minimum: 0 title: Step description: >- Step/Epoch number in an iterative training loop the model version was created. Generally useful when logging a model version from a MLRepo Run default: 0 run_id: type: string title: Run Id description: ID of the MLRepo Run that generated the model version type: object required: - name - metadata - ml_repo - source title: ModelManifest description: Model Version artifact. x-fern-type-name: ModelManifest ChatPromptManifest: properties: name: type: string maxLength: 256 pattern: ^[A-Za-z0-9_\-]+$ title: Name description: >- Name of the prompt (alphanumeric characters, hyphens, and underscores only, max 256 characters) metadata: type: object title: Metadata description: >- Key value metadata. Should be valid JSON. For e.g. `{"business-unit": "sales", "quality": "good", "rating": 4.5}` ml_repo: type: string pattern: ^[a-zA-Z][a-zA-Z0-9\-]{1,98}[a-zA-Z0-9]$ title: Ml Repo description: >- Name of the ML Repo that this prompt belongs to (must start and end with alphanumeric, 2-100 characters) version: type: integer minimum: 1 title: Version description: Version of the entity type: type: string enum: - chat_prompt title: Type default: chat_prompt description: type: string maxLength: 512 title: Description version_alias: type: string maxLength: 128 pattern: ^v[a-zA-Z0-9.-]*([a-zA-Z0-9]+)$ title: Version Alias description: >- Version alias is alternate, ideally human readable, version string to reference an artifact version. It should start with `v` followed by alphanumeric and it can include `.` and `-` in between (e.g. `v1.0.0`, `v1-prod`, `v3-dev`, etc) messages: items: anyOf: - $ref: '#/components/schemas/SystemMessage' - $ref: '#/components/schemas/AssistantMessage' - $ref: '#/components/schemas/UserMessage' - $ref: '#/components/schemas/ToolMessage' - $ref: '#/components/schemas/DeveloperMessage' type: array title: Messages description: List of messages in the chat conversation, must be non-empty variables: additionalProperties: type: string type: object title: Variables description: >- Variables referenced in messages and that can be replaced when running generation model_configuration: $ref: '#/components/schemas/ModelConfiguration' tools: items: $ref: '#/components/schemas/ToolSchema' type: array title: Tools description: List of tools to be used in the chat prompt mcp_servers: items: anyOf: - $ref: '#/components/schemas/MCPServerWithFQN' - $ref: '#/components/schemas/MCPServerWithURL' type: array title: Mcp Servers description: A list of MCP servers FQNs or URLs and their tools guardrails: $ref: '#/components/schemas/Guardrails' response_format: oneOf: - $ref: '#/components/schemas/JsonObjectResponseFormat' - $ref: '#/components/schemas/JsonSchemaResponseFormat' title: Response Format description: Response format configuration for structured outputs discriminator: propertyName: type mapping: json_object: $ref: '#/components/schemas/JsonObjectResponseFormat' json_schema: $ref: '#/components/schemas/JsonSchemaResponseFormat' routing_config: oneOf: - $ref: '#/components/schemas/WeightBasedLoadBalancing' - $ref: '#/components/schemas/LatencyBasedLoadBalancing' - $ref: '#/components/schemas/PriorityBasedLoadBalancing' title: Routing Config description: Configuration for routing requests to different model targets discriminator: propertyName: type mapping: weight-based-routing: $ref: '#/components/schemas/WeightBasedLoadBalancing' latency-based-routing: $ref: '#/components/schemas/LatencyBasedLoadBalancing' priority-based-routing: $ref: '#/components/schemas/PriorityBasedLoadBalancing' cache_config: oneOf: - $ref: '#/components/schemas/SemanticCacheConfig' - $ref: '#/components/schemas/ExactMatchCacheConfig' title: Cache Config description: Cache configuration for the chat prompt discriminator: propertyName: type mapping: semantic: $ref: '#/components/schemas/SemanticCacheConfig' exact-match: $ref: '#/components/schemas/ExactMatchCacheConfig' tool_call_to_mcp_mapping: additionalProperties: $ref: '#/components/schemas/MCPServerToolDetails' type: object title: Tool Call To Mcp Mapping description: Mapping of tool calls to MCP server integration IDs and tool names logging_config: $ref: '#/components/schemas/LoggingConfig' sub_agents: items: $ref: '#/components/schemas/SubAgent' type: array title: Sub Agents description: Sub agents to be used in the chat prompt type: object required: - name - metadata - ml_repo - messages title: ChatPromptManifest description: Chat Prompt manifest. x-fern-type-name: ChatPromptManifest ArtifactManifest: properties: name: type: string maxLength: 256 pattern: ^[A-Za-z0-9_\-]+$ title: Name description: >- Name of the artifact (alphanumeric characters, hyphens, and underscores only, max 256 characters) metadata: type: object title: Metadata description: >- Key value metadata. Should be valid JSON. For e.g. `{"business-unit": "sales", "quality": "good", "rating": 4.5}` ml_repo: type: string pattern: ^[a-zA-Z][a-zA-Z0-9\-]{1,98}[a-zA-Z0-9]$ title: Ml Repo description: >- Name of the ML Repo that this artifact belongs to (must start and end with alphanumeric, 2-100 characters) version: type: integer minimum: 1 title: Version description: Version of the entity type: type: string enum: - artifact-version title: Type default: artifact-version description: type: string maxLength: 512 title: Description version_alias: type: string maxLength: 128 pattern: ^v[a-zA-Z0-9.-]*([a-zA-Z0-9]+)$ title: Version Alias description: >- Version alias is alternate, ideally human readable, version string to reference an artifact version. It should start with `v` followed by alphanumeric and it can include `.` and `-` in between (e.g. `v1.0.0`, `v1-prod`, `v3-dev`, etc) source: oneOf: - $ref: '#/components/schemas/TrueFoundryManagedSource' - $ref: '#/components/schemas/ExternalBlobStorageSource' - $ref: '#/components/schemas/LocalArtifactSource' title: Source discriminator: propertyName: type mapping: truefoundry: $ref: '#/components/schemas/TrueFoundryManagedSource' external: $ref: '#/components/schemas/ExternalBlobStorageSource' local: $ref: '#/components/schemas/LocalArtifactSource' step: type: integer minimum: 0 title: Step description: >- Step/Epoch number in an iterative training loop the artifact version was created. Generally useful when logging a model version from a MLRepo Run default: 0 run_id: type: string title: Run Id description: ID of the MLRepo Run that generated the artifact version type: object required: - name - metadata - ml_repo - source title: ArtifactManifest description: Artifact Version manifest. x-fern-type-name: ArtifactManifest AgentSkillManifest: properties: name: type: string maxLength: 64 minLength: 1 pattern: ^(?!.*anthropic)(?!.*claude)[a-z0-9](?:[a-z0-9-]*[a-z0-9])?$ title: Name description: Name of the skill. metadata: type: object title: Metadata description: >- Key value metadata. Should be valid JSON. For e.g. `{"business-unit": "sales", "quality": "good", "rating": 4.5}` ml_repo: type: string pattern: ^[a-zA-Z][a-zA-Z0-9\-]{1,98}[a-zA-Z0-9]$ title: Ml Repo description: Name of the ML Repo version: type: integer minimum: 1 title: Version description: Version of the entity type: type: string enum: - agent-skill title: Type default: agent-skill source: oneOf: - $ref: '#/components/schemas/AgentSkillSourceInline' - $ref: '#/components/schemas/AgentSkillSourceBlobStorage' - $ref: '#/components/schemas/AgentSkillSourceLocal' title: Source discriminator: propertyName: type mapping: inline: $ref: '#/components/schemas/AgentSkillSourceInline' blob-storage: $ref: '#/components/schemas/AgentSkillSourceBlobStorage' local: $ref: '#/components/schemas/AgentSkillSourceLocal' type: object required: - name - metadata - ml_repo - source title: AgentSkillManifest description: >- Apply/API agent-skill manifest; ``source`` is inline (apply) or blob-storage (stored). x-fern-type-name: AgentSkillManifest TrueFoundryManagedSource: properties: type: type: string enum: - truefoundry title: Type description: Type of the source uri: type: string pattern: ^(s3|gs|wasbs)://.* title: Uri type: object required: - type title: TrueFoundryManagedSource x-fern-type-name: TrueFoundryManagedSource ExternalBlobStorageSource: properties: type: type: string enum: - external title: Type description: Type of the source uri: type: string pattern: ^(s3|gs|wasbs)://.* title: Uri description: >- URI referencing a path in the blob storage bucket linked to the MLRepo type: object required: - type - uri title: ExternalBlobStorageSource x-fern-type-name: ExternalBlobStorageSource LocalModelSource: properties: type: type: string enum: - local title: Type description: Type of the source file_or_folder: type: string title: File Or Folder description: Path to the model file or folder type: object required: - type - file_or_folder title: LocalModelSource x-fern-type-name: LocalModelSource TransformersFramework: properties: type: type: string enum: - transformers title: Type description: Type of the framework library_name: allOf: - $ref: '#/components/schemas/LibraryName' description: Name of the library for the framework default: transformers pipeline_tag: type: string title: Pipeline Tag description: >- The `pipeline()` task this model can be used with e.g. `text-generation`. See [huggingface docs](https://huggingface.co/docs/transformers/main/en/main_classes/pipelines#transformers.pipeline.task) for all possible values base_model: type: string title: Base Model description: >- Base model Id from HuggingFace Hub. If this is a finetuned model, this points to the base model id used for finetuning. type: object required: - type title: TransformersFramework x-fern-type-name: TransformersFramework TensorFlowFramework: properties: type: type: string enum: - tensorflow title: Type description: Type of the framework type: object required: - type title: TensorFlowFramework x-fern-type-name: TensorFlowFramework SklearnFramework: properties: type: type: string enum: - sklearn title: Type description: Type of the framework model_filepath: type: string title: Model Filepath description: Relative path to the model file in the model version contents serialization_format: $ref: '#/components/schemas/SklearnSerializationFormat' model_schema: $ref: '#/components/schemas/SklearnModelSchema' type: object required: - type title: SklearnFramework x-fern-type-name: SklearnFramework PyTorchFramework: properties: type: type: string enum: - pytorch title: Type description: Type of the framework type: object required: - type title: PyTorchFramework x-fern-type-name: PyTorchFramework KerasFramework: properties: type: type: string enum: - keras title: Type description: Type of the framework type: object required: - type title: KerasFramework x-fern-type-name: KerasFramework XGBoostFramework: properties: type: type: string enum: - xgboost title: Type description: Type of the framework serialization_format: $ref: '#/components/schemas/XGBoostSerializationFormat' model_filepath: type: string title: Model Filepath description: Relative path to the model file in the model version contents model_schema: $ref: '#/components/schemas/XGBoostModelSchema' type: object required: - type title: XGBoostFramework x-fern-type-name: XGBoostFramework LightGBMFramework: properties: type: type: string enum: - lightgbm title: Type description: Type of the framework type: object required: - type title: LightGBMFramework x-fern-type-name: LightGBMFramework FastAIFramework: properties: type: type: string enum: - fastai title: Type description: Type of the framework type: object required: - type title: FastAIFramework x-fern-type-name: FastAIFramework H2OFramework: properties: type: type: string enum: - h2o title: Type description: Type of the framework type: object required: - type title: H2OFramework x-fern-type-name: H2OFramework ONNXFramework: properties: type: type: string enum: - onnx title: Type description: Type of the framework type: object required: - type title: ONNXFramework x-fern-type-name: ONNXFramework SpaCyFramework: properties: type: type: string enum: - spacy title: Type description: Type of the framework type: object required: - type title: SpaCyFramework x-fern-type-name: SpaCyFramework StatsModelsFramework: properties: type: type: string enum: - statsmodels title: Type description: Type of the framework type: object required: - type title: StatsModelsFramework x-fern-type-name: StatsModelsFramework GluonFramework: properties: type: type: string enum: - gluon title: Type description: Type of the framework type: object required: - type title: GluonFramework x-fern-type-name: GluonFramework PaddleFramework: properties: type: type: string enum: - paddle title: Type description: Type of the framework type: object required: - type title: PaddleFramework x-fern-type-name: PaddleFramework ModelVersionEnvironment: properties: python_version: type: string pattern: ^\d+(\.\d+){1,2}([\-\.a-z0-9]+)?$ title: Python Version description: Python version to use when running the model pip_packages: items: type: string type: array title: Pip Packages description: pip dependencies needed for running the model type: object title: ModelVersionEnvironment description: >- Runtime environment describing python version and dependencies for running the model x-fern-type-name: ModelVersionEnvironment SystemMessage: description: System message for the chat type: object required: - role - content properties: role: description: Role of the message type: string enum: - system content: description: Text content for the system message oneOf: - type: string pattern: ^.[\s\S]*$ - $ref: '#/components/schemas/BlobStorageReference' name: description: Name of the system type: string title: SystemMessage x-fern-type-name: SystemMessage AssistantMessage: description: Assistant message turn type: object required: - role properties: role: description: Role of the message type: string enum: - assistant content: description: Text content for the assistant message oneOf: - type: string - type: array items: type: object oneOf: - $ref: '#/components/schemas/TextContentPart' - $ref: '#/components/schemas/RefusalContentPart' - $ref: '#/components/schemas/BlobStorageReference' tool_calls: description: List of tool calls made by the assistant type: array items: $ref: '#/components/schemas/ToolCall' name: description: Name of the assistant this message is from type: string title: AssistantMessage x-fern-type-name: AssistantMessage UserMessage: description: User message turn type: object required: - role - content properties: role: description: Role of the message type: string enum: - user content: description: Content of the user message. can be a mix of text and images oneOf: - type: string pattern: ^.[\s\S]*$ - type: array items: type: object oneOf: - $ref: '#/components/schemas/TextContentPart' - $ref: '#/components/schemas/ImageContentPart' - $ref: '#/components/schemas/BlobStorageReference' name: description: Name of the user this message is from type: string title: UserMessage x-fern-type-name: UserMessage ToolMessage: description: Tool message for the chat type: object required: - role - content - tool_call_id properties: role: description: Role of the message type: string enum: - tool content: description: Content of the tool call result oneOf: - type: string pattern: ^.[\s\S]*$ - $ref: '#/components/schemas/BlobStorageReference' tool_call_id: description: Unique identifier for the tool call type: string title: ToolMessage x-fern-type-name: ToolMessage DeveloperMessage: description: Developer message for the chat type: object required: - role - content properties: role: description: The role of the messages author, in this case developer. type: string enum: - developer content: description: The contents of the developer message. oneOf: - type: string pattern: ^.[\s\S]*$ - $ref: '#/components/schemas/BlobStorageReference' name: description: An optional name for the participant. type: string title: DeveloperMessage x-fern-type-name: DeveloperMessage ModelConfiguration: properties: provider: type: string title: Provider description: >- Name of the provider, must be one of the integration providers configured for the Gateway model: type: string pattern: ^[a-zA-Z](?:[a-zA-Z0-9\-.]{0,60}[a-zA-Z0-9])$ title: Model description: >- Name of the model to be used for generations. This model should be available in the provider parameters: allOf: - $ref: '#/components/schemas/Parameters' title: Parameters description: Parameters to pass to the model when generating extra_parameters: type: object title: Extra Parameters description: Arbitrary extra parameters to pass to the model when generating type: object required: - provider - model title: ModelConfiguration description: >- Add a model from Gateway along with parameters to be used for chat completion x-fern-type-name: ModelConfiguration ToolSchema: description: Schema defining a tool for the chat prompt type: object required: - type - function properties: type: description: Type of the tool type: string enum: - function function: $ref: '#/components/schemas/FunctionSchema' title: ToolSchema x-fern-type-name: ToolSchema MCPServerWithFQN: description: MCP server with FQN type: object required: - type - integration_fqn - enable_all_tools properties: type: description: The FQN of the MCP server type: string enum: - mcp-server-fqn integration_fqn: description: The FQN of the MCP server type: string enable_all_tools: description: Whether to enable all tools from the MCP server type: boolean default: false tools: description: The tools to enable from the MCP server type: array items: $ref: '#/components/schemas/MCPTool' title: MCPServerWithFQN x-fern-type-name: MCPServerWithFQN MCPServerWithURL: description: MCP server with URL type: object required: - type - url - enable_all_tools properties: type: description: The URL of the MCP server type: string enum: - mcp-server-url url: description: The URL of the MCP server type: string pattern: ^https?://.+ headers: description: The headers to send to the MCP server type: object additionalProperties: type: string enable_all_tools: description: Whether to enable all tools from the MCP server type: boolean default: false tools: description: The tools to enable from the MCP server type: array items: $ref: '#/components/schemas/MCPTool' title: MCPServerWithURL x-fern-type-name: MCPServerWithURL Guardrails: description: Guardrails type: object properties: input_guardrails: description: >- List of guardrail names to apply to user input messages before processing. type: array items: type: string output_guardrails: description: >- List of guardrail names to apply to AI response messages before returning to user. type: array items: type: string title: Guardrails x-fern-type-name: Guardrails JsonObjectResponseFormat: description: Simple JSON mode response format type: object required: - type properties: type: description: Type of response format type: string enum: - json_object title: JsonObjectResponseFormat x-fern-type-name: JsonObjectResponseFormat JsonSchemaResponseFormat: description: Structured JSON response format with schema validation type: object required: - type - json_schema properties: type: description: Type of response format type: string enum: - json_schema json_schema: $ref: '#/components/schemas/JsonSchema' title: JsonSchemaResponseFormat x-fern-type-name: JsonSchemaResponseFormat WeightBasedLoadBalancing: description: Weight-based Load Balancing type: object required: - type - load_balance_targets properties: type: description: Routing Type type: string enum: - weight-based-routing load_balance_targets: description: List of targets for load balancing with weights type: array items: $ref: '#/components/schemas/LoadBalanceTarget' sticky_routing: $ref: '#/components/schemas/StickyRouting' title: WeightBasedLoadBalancing x-fern-type-name: WeightBasedLoadBalancing LatencyBasedLoadBalancing: description: Latency-based Load Balancing type: object required: - type - load_balance_targets properties: type: description: Routing Type type: string enum: - latency-based-routing load_balance_targets: description: List of targets for latency-based load balancing type: array items: $ref: '#/components/schemas/LatencyBasedLoadBalanceTarget' title: LatencyBasedLoadBalancing x-fern-type-name: LatencyBasedLoadBalancing PriorityBasedLoadBalancing: description: Priority-based Load Balancing type: object required: - type - load_balance_targets properties: type: description: Routing Type type: string enum: - priority-based-routing load_balance_targets: description: List of targets for priority-based load balancing type: array items: $ref: '#/components/schemas/PriorityBasedLoadBalanceTarget' title: PriorityBasedLoadBalancing x-fern-type-name: PriorityBasedLoadBalancing SemanticCacheConfig: description: Semantic Cache Configuration type: object required: - type - similarity_threshold - ttl properties: type: description: Cache type type: string enum: - semantic similarity_threshold: description: Similarity threshold for semantic matching (between 0 and 1) type: number minimum: 0 exclusiveMinimum: true maximum: 1 namespace: description: Cache namespace (defaults to 'default' if not provided) type: string ttl: description: Time-to-live for cached entries in seconds (max 3 days) type: number minimum: 0 exclusiveMinimum: true maximum: 259200 default: 3600 title: SemanticCacheConfig x-fern-type-name: SemanticCacheConfig ExactMatchCacheConfig: description: Exact Match Cache Configuration type: object required: - type - ttl properties: type: description: Cache type type: string enum: - exact-match namespace: description: Cache namespace (defaults to 'default' if not provided) type: string ttl: description: Time-to-live for cached entries in seconds (max 3 days) type: number minimum: 0 exclusiveMinimum: true maximum: 259200 default: 3600 title: ExactMatchCacheConfig x-fern-type-name: ExactMatchCacheConfig MCPServerToolDetails: description: >- Mapping configuration for tool calls to MCP server integration and tool names type: object required: - mcp_server_integration_id - tool_name properties: mcp_server_integration_id: description: The integration ID of the MCP server type: string tool_name: description: The name of the tool in the MCP server type: string title: MCPServerToolDetails x-fern-type-name: MCPServerToolDetails LoggingConfig: description: Logging configuration for the chat prompt type: object required: - enabled properties: enabled: description: Whether logging is enabled for the chat prompt type: boolean default: true tracing_project_fqn: description: The FQN of the tracing project where the logs will be stored type: string title: LoggingConfig x-fern-type-name: LoggingConfig SubAgent: description: Sub agent configuration for the chat prompt type: object required: - name properties: name: description: Name of the sub agent type: string title: SubAgent x-fern-type-name: SubAgent LocalArtifactSource: properties: type: type: string enum: - local title: Type description: Type of the source paths: items: $ref: '#/components/schemas/ArtifactPath' type: array title: Paths description: >- Array of ArtifactPath objects representing the source and destination paths type: object required: - type - paths title: LocalArtifactSource x-fern-type-name: LocalArtifactSource AgentSkillSourceInline: properties: type: type: string enum: - inline title: Type skill_md: type: string maxLength: 20000 minLength: 1 title: Skill Md description: Main body of SKILL.md file. type: object required: - type - skill_md title: AgentSkillSourceInline description: SKILL.md for this version when using inline source. x-fern-type-name: AgentSkillSourceInline AgentSkillSourceBlobStorage: properties: type: type: string enum: - blob-storage title: Type description: type: string maxLength: 1024 minLength: 1 title: Description description: 1–1024 chars. uri: type: string minLength: 1 title: Uri description: Storage URI of the blob storage object where the skill is stored. type: object required: - type - description - uri title: AgentSkillSourceBlobStorage description: SKILL.md content stored on blob x-fern-type-name: AgentSkillSourceBlobStorage AgentSkillSourceLocal: properties: type: type: string enum: - local title: Type skill_dir: type: string minLength: 1 title: Skill Dir description: >- Path to the skill root directory, including SKILL.md and related assets. type: object required: - type - skill_dir title: AgentSkillSourceLocal description: >- Load the skill from a local directory. The directory must include SKILL.md with valid YAML front matter. x-fern-type-name: AgentSkillSourceLocal LibraryName: type: string enum: - transformers - sentence-transformers - diffusers title: LibraryName description: Name of the library for the framework x-fern-type-name: LibraryName SklearnSerializationFormat: type: string enum: - cloudpickle - joblib - pickle title: SklearnSerializationFormat description: Serialization format used to save the sklearn model x-fern-type-name: SklearnSerializationFormat SklearnModelSchema: properties: infer_method_name: allOf: - $ref: '#/components/schemas/InferMethodName' description: Name of the method used for inference inputs: items: type: object type: array title: Inputs description: Schema of the input outputs: items: type: object type: array title: Outputs description: Schema of the output type: object required: - infer_method_name - inputs - outputs title: SklearnModelSchema x-fern-type-name: SklearnModelSchema XGBoostSerializationFormat: type: string enum: - cloudpickle - joblib - pickle - json title: XGBoostSerializationFormat description: Serialization format used to save the xgboost model x-fern-type-name: XGBoostSerializationFormat XGBoostModelSchema: properties: infer_method_name: type: string enum: - predict title: Infer Method Name description: Name of the method used for inference inputs: items: type: object type: array title: Inputs description: Schema of the input outputs: items: type: object type: array title: Outputs description: Schema of the output type: object required: - infer_method_name - inputs - outputs title: XGBoostModelSchema description: Schema of the XGBoost model x-fern-type-name: XGBoostModelSchema BlobStorageReference: properties: type: type: string enum: - blob-storage title: Type path: type: string pattern: ^.{1,}$ title: Path description: Path to the content in blob storage mime_type: allOf: - $ref: '#/components/schemas/MimeType' description: MIME type of the content type: object required: - type - path - mime_type title: BlobStorageReference description: A reference to content offloaded to blob storage x-fern-type-name: BlobStorageReference TextContentPart: description: Text content for the message type: object required: - type - text properties: type: description: Type of the content part type: string enum: - text text: description: Text content for the message oneOf: - type: string pattern: ^.[\s\S]*$ - $ref: '#/components/schemas/BlobStorageReference' title: TextContentPart x-fern-type-name: TextContentPart RefusalContentPart: type: object required: - type - refusal properties: type: description: Type of the content part type: string refusal: description: Reason type: string title: RefusalContentPart x-fern-type-name: RefusalContentPart ToolCall: type: object required: - type - id - function properties: type: description: Type of the tool call type: string enum: - function id: description: Unique identifier for the tool call type: string function: $ref: '#/components/schemas/Function' title: ToolCall x-fern-type-name: ToolCall ImageContentPart: description: Image content for the message type: object required: - type - image_url properties: type: description: Type of the content part type: string enum: - image_url image_url: description: Image URL linking to the image type: object required: - url properties: url: description: Image URL linking to the image oneOf: - type: string pattern: "^\b((https?://)?(?:www\\.)?[-a-zA-Z0-9@:%._\\+~#=]{1,256}\\.[a-zA-Z0-9()]{1,6}(?:[-a-zA-Z0-9()@:%_\\+.~#?&/=]*)|data:image/[a-zA-Z]+;base64,[a-zA-Z0-9+/=\\s]+)$" - $ref: '#/components/schemas/BlobStorageReference' detail: description: Details type: string title: ImageContentPart x-fern-type-name: ImageContentPart Parameters: properties: max_tokens: type: integer title: Max Tokens temperature: type: number title: Temperature top_k: type: number title: Top K top_p: type: number title: Top P stop: anyOf: - type: array items: type: string - type: string title: Stop type: object title: Parameters description: Parameters to pass to the model when generating x-fern-type-name: Parameters FunctionSchema: description: Schema defining a function for tool calls type: object required: - name properties: name: description: Name of the function type: string description: description: Description of the function type: string parameters: description: Parameters schema for the function type: object strict: description: Indicates if the function should be called strictly type: boolean default: false title: FunctionSchema x-fern-type-name: FunctionSchema MCPTool: description: Tool configuration for MCP server type: object required: - name properties: name: description: The name of the tool as it appears in the MCP server type: string title: MCPTool x-fern-type-name: MCPTool JsonSchema: description: JSON schema for structured output type: object required: - name - schema properties: name: description: Name of the schema type: string schema: description: JSON schema object defining the expected structure type: object title: JsonSchema x-fern-type-name: JsonSchema LoadBalanceTarget: description: Weight-based Target type: object required: - target - weight properties: target: description: Target model or provider FQN type: string weight: description: >- Relative weight for routing decisions (higher values mean more traffic) type: integer retry_config: $ref: '#/components/schemas/RetryConfig' fallback_status_codes: description: >- Status Codes for which the request will fallback to other targets. If the status code is not present in fallback_status_codes, it fails immediately. type: array items: type: string default: - '401' - '403' - '404' - '408' - '429' - '500' - '502' - '503' fallback_candidate: description: >- Whether this target is a fallback candidate. If set to false, this model will not be considered as a fallback option for targets of this load-balance-rule type: boolean default: true override_params: description: Optional parameters to override in the request type: object headers_override: $ref: '#/components/schemas/HeadersOverride' metadata_match: description: >- Optional metadata key-value pairs that must match incoming request metadata headers for this target to be considered for routing. type: object additionalProperties: type: string title: LoadBalanceTarget x-fern-type-name: LoadBalanceTarget StickyRouting: description: Sticky Routing type: object required: - ttl_seconds - session_identifiers properties: ttl_seconds: description: Sticky mapping TTL in seconds type: integer minimum: 1800 maximum: 86400 session_identifiers: description: List of identifier selectors used to build sticky key type: array items: $ref: '#/components/schemas/StickySessionIdentifier' title: StickyRouting x-fern-type-name: StickyRouting LatencyBasedLoadBalanceTarget: description: Latency-based Target type: object required: - target properties: target: description: Target model or provider FQN type: string retry_config: $ref: '#/components/schemas/RetryConfig' fallback_status_codes: description: >- Status Codes for which the request will fallback to other targets. If the status code is not present in fallback_status_codes, it fails immediately. type: array items: type: string default: - '401' - '403' - '404' - '408' - '429' - '500' - '502' - '503' fallback_candidate: description: >- Whether this target is a fallback candidate. If set to false, this model will not be considered as a fallback option for targets of this load-balance-rule type: boolean default: true override_params: description: Optional parameters to override in the request type: object headers_override: $ref: '#/components/schemas/HeadersOverride' metadata_match: description: >- Optional metadata key-value pairs that must match incoming request metadata headers for this target to be considered for routing. type: object additionalProperties: type: string title: LatencyBasedLoadBalanceTarget x-fern-type-name: LatencyBasedLoadBalanceTarget PriorityBasedLoadBalanceTarget: type: object required: - target - priority properties: target: description: Target model or provider FQN type: string priority: description: >- Priority for the target, Lower the number, higher the priority (0 is the highest priority) type: integer minimum: 0 default: 0 sla_cutoff: $ref: '#/components/schemas/SLACutoff' retry_config: $ref: '#/components/schemas/RetryConfig' fallback_status_codes: description: >- Status Codes for which the request will fallback to other targets. If the status code is not present in fallback_status_codes, it fails immediately. type: array items: type: string default: - '401' - '403' - '404' - '408' - '429' - '500' - '502' - '503' fallback_candidate: description: >- Whether this target is a fallback candidate. If set to false, this model will not be considered as a fallback option for targets of this load-balance-rule type: boolean default: true override_params: description: Optional parameters to override in the request type: object headers_override: $ref: '#/components/schemas/HeadersOverride' metadata_match: description: >- Optional metadata key-value pairs that must match incoming request metadata headers for this target to be considered for routing. type: object additionalProperties: type: string title: PriorityBasedLoadBalanceTarget x-fern-type-name: PriorityBasedLoadBalanceTarget ArtifactPath: properties: src: type: string title: Src description: Local file or folder path dest: type: string title: Dest description: >- Relative path where the file or folder will be uploaded to in the artifact type: object required: - src title: ArtifactPath x-fern-type-name: ArtifactPath InferMethodName: type: string enum: - predict - predict_proba title: InferMethodName description: Name of the method used for inference x-fern-type-name: InferMethodName MimeType: type: string enum: - text/plain - application/json - image/png - image/jpeg - application/x-directory title: MimeType description: MIME type of the content x-fern-type-name: MimeType Function: type: object required: - name - arguments properties: name: description: Name of the tool/function being called type: string arguments: description: Arguments passed to the tool/function type: string title: Function x-fern-type-name: Function RetryConfig: type: object required: - attempts properties: attempts: description: Number of attempts to retry the request type: integer minimum: 0 delay: description: Delay between retries in milliseconds type: integer minimum: 0 exclusiveMinimum: true default: 100 on_status_codes: description: Status codes to retry on type: array items: type: string default: - '429' - '500' - '502' - '503' title: RetryConfig x-fern-type-name: RetryConfig HeadersOverride: description: Headers Override type: object properties: remove: description: List of headers to remove from upstream request type: array items: type: string set: description: Headers to add or overwrite in upstream request type: object additionalProperties: type: string title: HeadersOverride x-fern-type-name: HeadersOverride StickySessionIdentifier: description: Sticky Session Identifier type: object required: - source - key properties: source: description: Source from which session identifier value will be extracted type: string enum: - headers - metadata key: description: Key to read from selected source type: string pattern: ^[!#$%&'*+.^_`|~0-9A-Za-z-]{1,128}$ title: StickySessionIdentifier x-fern-type-name: StickySessionIdentifier SLACutoff: type: object properties: time_per_output_token_ms: description: >- Time per output token threshold in milliseconds. If time per output token exceeds this value, the target will be marked as unhealthy type: integer minimum: 0 exclusiveMinimum: true time_to_first_token_ms: description: >- Time to first token threshold in milliseconds. If time to first token exceeds this value, the target will be marked as unhealthy type: integer minimum: 0 exclusiveMinimum: true title: SLACutoff x-fern-type-name: SLACutoff securitySchemes: jwt: type: http scheme: bearer bearerFormat: JWT ````