gen_ai_hub.prompt_registry package

get_orchestration_configs(scenario, name, version, retrieve=None, include_spec=None, resolve_template_ref=None)

Retrieve the latest version of every orchestration config based on the filters.

Parameters:

scenario (str) -- the scenario name of the orchestration config.
name (str) -- the name of the orchestration config.
version (str) -- the version of the orchestration config.
retrieve (str, optional) -- both(default), imperative, declarative
include_spec (bool, optional) -- false(default), true
resolve_template_ref (bool, optional) -- false(default), true

Returns:

An OrchestrationConfigListResponse object.

Return type:

https://api.sap.com/api/PROMPT_REGISTRY_API/overview

import_orchestration_config(file)

Import a runtime/declarative orchestration config into the design time environment.

Parameters:: file (bytes) -- binary file content
Returns:: A OrchestrationConfigPostResponse object.
Return type:: OrchestrationConfigPostResponse

class OrchestrationConfigDeleteResponse

Bases: BaseModel

Response to a delete orchestration config request.

Args:: message: Response message.

message: str

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class OrchestrationConfigGetResponse

Bases: BaseModel

Response to a get orchestration config request.

Args:: id: UUID of the config. name: Config name. version: Config version. scenario: Scenario name. creation_timestamp: When the config was created. managed_by: Who manages the config. is_version_head: Whether this is the head version. spec: The orchestration config specification (optional).

model_dump(**kwargs): Dumps the model to a dictionary with default settings.

creation_timestamp: str | None

id: str | None

is_version_head: bool | None

managed_by: str | None

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str | None

resource_group_id: str | None

scenario: str | None

spec: OrchestrationConfig | None

version: str | None

class OrchestrationConfigListResponse

Bases: BaseModel

Response to list orchestration configs request.

Args:: count: Number of configs returned. resources: List of OrchestrationConfigGetResponse objects.

model_dump(**kwargs): Dumps the model to a dictionary with default settings.

count: int

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

resources: List[OrchestrationConfigGetResponse]

class OrchestrationConfigPostRequest

Bases: BaseModel

Request to create an orchestration config.

Args:: name: The name of the orchestration config. version: The version of the orchestration config. scenario: The scenario of the orchestration config. spec: The orchestration config specification.

model_dump(**kwargs): Dumps the model to a dictionary with default settings.

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str

scenario: str

spec: OrchestrationConfig

version: str

class OrchestrationConfigPostResponse

Bases: BaseModel

Response to the orchestration config post request.

Args:: message: Response message. id: UUID of the created/updated config. scenario: The scenario name. name: The config name. version: The config version.

id: str

message: str

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str

scenario: str

version: str

class PromptTemplate

Bases: BaseModel

Represents a prompt template.

Args:

role: The role of the prompt template.

content: The content of the prompt template.

classmethod content_validation(content): Validates and maps the content field to the appropriate types.

content: str | List[str | TextPart | ImagePart | ImageItem]: The content of the prompt template.

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

role: str: The role of the prompt template.

class PromptTemplateClient

Bases: PromptRegistryClient

Client for interacting with the Prompt Registry Prompt Template API.

create_prompt_template(name, version, scenario, prompt_template_spec)

Create or update a prompt template.

Parameters:

name (str) -- the name of the prompt template.
version (str) -- the version of the prompt template.
scenario (str) -- the scenario name of the prompt template.
prompt_template_spec (PromptTemplateSpec) -- the specification of the prompt template.

Returns:

A PromptTemplatePostResponse object.

Return type:

PromptTemplatePostResponse

delete_prompt_template_by_id(template_id)

Delete a specific version of the prompt template by ID.

Parameters:: template_id (str) -- The ID of the prompt template to delete.
Returns:: A PromptTemplateDeleteResponse object.
Return type:: PromptTemplateDeleteResponse

export_prompt_template(template_id)

Export a design time template in a declarative compatible yaml file. Supports only single file export.

Parameters:: template_id (str) -- The id of the prompt template to export.
Returns:: bytes: The content of the exported file
Return type:: bytes

fill_prompt_template(scenario, name, version, input_params, metadata=False)

Replace the placeholders of the prompt template referenced via scenario-name-version with user provided values.

Parameters:

scenario (str) -- the scenario name of the prompt template.
name (str) -- the name of the prompt template.
version (str) -- the version of the prompt template.
input_params (dict) -- User provided values to replace the placeholders of the prompt template.
metadata (bool, optional) -- False(default), True return resource object with all details.

Returns:

A PromptTemplateSubstitutionResponse object.

Return type:

fill_prompt_template_by_id(template_id, input_params, metadata=False)

Replace the placeholders of the prompt template referenced via template_id with user provided values.

Parameters:

template_id (str) -- The ID of the prompt template.
input_params (dict) -- User provided values to replace the placeholders of the prompt template.
metadata (bool, optional) -- False(default), True return resource object with all details.

Returns:

A PromptTemplateSubstitutionResponse object.

Return type:

get_prompt_template_by_id(template_id)

Retrieve a specific version of the prompt template by ID.

Parameters:: template_id (str) -- The ID of the prompt template to retrieve.
Returns:: A PromptTemplateGetResponse object.
Return type:: PromptTemplateGetResponse

get_prompt_template_history(scenario, name, version)

Retrieve the history of edits to the prompt template. Only for imperative managed prompt templates.

Parameters:

scenario (str) -- The scenario name of the prompt template.
name (str) -- The name of the prompt template.
version (str) -- The version ID of the prompt template.

Returns:

A PromptTemplateListResponse object.

Return type:

get_prompt_templates(scenario, name, version, retrieve=None, include_spec=None)

Retrieve the latest version of every prompt template based on the filters.

Parameters:

scenario (str) -- the scenario name of the prompt template.
name (str) -- the name of the prompt template.
version (str) -- the version of the prompt template.
retrieve (str, optional) -- both(default), imperative, declarative
include_spec (bool, optional) -- false(default), true

Returns:

A PromptTemplateListResponse object.

Return type:

https://api.sap.com/api/PROMPT_REGISTRY_API/overview

import_prompt_template(file)

Import a runtime/declarative prompt template into the design time environment.

Parameters:: file (bytes) -- binary file content
Returns:: A PromptTemplatePostResponse object.
Return type:: PromptTemplatePostResponse

class PromptTemplateDeleteResponse

Bases: BaseModel

Represents a response to a request to delete a prompt template.

Args:: message: The message of the response.

message: str: The message of the response.

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class PromptTemplateGetResponse

Bases: BaseModel

Represents a response to a request to get a prompt template.

Args:

id: The ID of the prompt template.

version: The version of the prompt template.

scenario: The scenario of the prompt template.

creation_timestamp: The creation timestamp of the prompt template.

managed_by: The manager of the prompt template.

is_version_head: Whether the version is the head version.

spec: The specification of the prompt template.

creation_timestamp: str | None: The creation timestamp of the prompt template.

id: str: The ID of the prompt template.

is_version_head: bool | None: Whether the version is the head version.

managed_by: str | None: The manager of the prompt template.

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str: The name of the prompt template.

scenario: str: The scenario of the prompt template.

spec: PromptTemplateSpec | None: The specification of the prompt template.

version: str: The version of the prompt template.

class PromptTemplateListResponse

Bases: BaseModel

Represents a response to a request to list prompt templates.

Args:

resources: The list of PromptGetResponse objects.

count: int: The number of prompt templates.

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

resources: List[PromptTemplateGetResponse]: The list of PromptGetResponse objects.

class PromptTemplatePostRequest

Bases: BaseModel

Represents a request to create a prompt template.

Args:

version: The version of the prompt template.

scenario: The scenario of the prompt template.

spec: The specification of the prompt template.

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str: The name of the prompt template.

scenario: str: The scenario of the prompt template.

spec: PromptTemplateSpec: The specification of the prompt template.

version: str: The version of the prompt template.

class PromptTemplatePostResponse

Bases: BaseModel

Represents a response to a request to create a prompt template.

Args:

message: The message of the response.

id: The ID of the prompt template.

scenario: The scenario of the prompt template.

version: The version of the prompt template.

id: str: The ID of the prompt template.

message: str: The message of the response.

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

name: str: The name of the prompt template.

scenario: str: The scenario of the prompt template.

version: str: The version of the prompt template.

class PromptTemplateSpec

Bases: BaseModel

Represents a prompt template specification.

Args:

Args: template: A list of prompt messages that form the template.

defaults: A dict of default values for template variables.

tools: A list of tool definitions.

response_format: A response format that the model output should adhere to.

additional_fields: Additional fields for the prompt template.

additional_fields: Dict[Any, Any] | None

defaults: dict | None

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

response_format: ResponseFormatText | ResponseFormatJsonObject | ResponseFormatJsonSchema | None

template: List[PromptTemplate]

tools: List[dict | FunctionTool] | None

class PromptTemplateSubstitutionRequest

Bases: BaseModel

Represents a request to substitute a prompt template.

Args:: input_params: User provided values to replace the placeholders of the prompt template.

input_params: Dict[Any, Any] | None: User provided values to replace the placeholders of the prompt template.

model_config: ClassVar[ConfigDict] = {'populate_by_name': True, 'validate_by_alias': True, 'validate_by_name': True}: Pydantic configuration to allow population by field name.

class PromptTemplateSubstitutionResponse

Bases: BaseModel

Represents a response to a request to substitute a prompt template.

Args:

parsed_prompt: The parsed prompt.

resource: List of TemplateGetResponse objects.

model_config: ClassVar[ConfigDict] = {}: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

parsed_prompt: List[PromptTemplate]: The parsed prompt.

resource: PromptTemplateGetResponse | None: List of TemplateGetResponse objects.

Subpackages

gen_ai_hub.prompt_registry.models package

Submodules

gen_ai_hub.prompt_registry.client module

class OrchestrationConfigClient

Bases: PromptRegistryClient

Client for interacting with the Prompt Registry Orchestration Config API.

create_orchestration_config(name, version, scenario, spec)

Create an orchestration config.

Parameters:

name (str) -- the name of the orchestration config.
version (str) -- the version of the orchestration config.
scenario (str) -- the scenario name of the orchestration config.
spec (Union[dict, OrchestrationConfig]) -- the specification of the orchestration config.

Returns:

An OrchestrationConfigPostResponse object.

Return type:

OrchestrationConfigPostResponse

delete_orchestration_config_by_id(config_id)

Delete a specific version of the orchestration config by ID.

Parameters:: config_id (str) -- The ID of the orchestration config.
Returns:: An OrchestrationConfigDeleteResponse object.
Return type:: OrchestrationConfigDeleteResponse

export_orchestration_config(config_id)

Export a design orchestration config in a declarative compatible yaml file. Supports only single file export.

Parameters:: config_id (str) -- The id of the orchestration config to export.
Returns:: bytes: The content of the exported file
Return type:: bytes

get_orchestration_config_by_id(config_id, resolve_template_ref=None)

Retrieve a specific version of the orchestration config by ID.

Parameters:

config_id (str) -- The ID of the orchestration config to retrieve.
resolve_template_ref (bool, optional) -- false(default), true

Returns:

An OrchestrationConfigGetResponse object.

Return type:

OrchestrationConfigGetResponse

get_orchestration_config_history(scenario, name, version, include_spec=None, resolve_template_ref=None)

Retrieve the history of edits to the orchestration config.

Parameters:

scenario (str) -- The scenario name of the orchestration config.
name (str) -- The name of the orchestration config.
version (str) -- The version ID of the orchestration config.
include_spec (bool, optional) -- false(default), true
resolve_template_ref (bool, optional) -- false(default), true

Returns:

An OrchestrationConfigListResponse object.

Return type:

get_orchestration_configs(scenario, name, version, retrieve=None, include_spec=None, resolve_template_ref=None)

Retrieve the latest version of every orchestration config based on the filters.

Parameters:

scenario (str) -- the scenario name of the orchestration config.
name (str) -- the name of the orchestration config.
version (str) -- the version of the orchestration config.
retrieve (str, optional) -- both(default), imperative, declarative
include_spec (bool, optional) -- false(default), true
resolve_template_ref (bool, optional) -- false(default), true

Returns:

An OrchestrationConfigListResponse object.

Return type:

https://api.sap.com/api/PROMPT_REGISTRY_API/overview

import_orchestration_config(file)

Import a runtime/declarative orchestration config into the design time environment.

Parameters:: file (bytes) -- binary file content
Returns:: A OrchestrationConfigPostResponse object.
Return type:: OrchestrationConfigPostResponse

class PromptTemplateClient

Bases: PromptRegistryClient

Client for interacting with the Prompt Registry Prompt Template API.

create_prompt_template(name, version, scenario, prompt_template_spec)

Create or update a prompt template.

Parameters:

name (str) -- the name of the prompt template.
version (str) -- the version of the prompt template.
scenario (str) -- the scenario name of the prompt template.
prompt_template_spec (PromptTemplateSpec) -- the specification of the prompt template.

Returns:

A PromptTemplatePostResponse object.

Return type:

PromptTemplatePostResponse

delete_prompt_template_by_id(template_id)

Delete a specific version of the prompt template by ID.

Parameters:: template_id (str) -- The ID of the prompt template to delete.
Returns:: A PromptTemplateDeleteResponse object.
Return type:: PromptTemplateDeleteResponse

export_prompt_template(template_id)

Export a design time template in a declarative compatible yaml file. Supports only single file export.

Parameters:: template_id (str) -- The id of the prompt template to export.
Returns:: bytes: The content of the exported file
Return type:: bytes

fill_prompt_template(scenario, name, version, input_params, metadata=False)

Replace the placeholders of the prompt template referenced via scenario-name-version with user provided values.

Parameters:

scenario (str) -- the scenario name of the prompt template.
name (str) -- the name of the prompt template.
version (str) -- the version of the prompt template.
input_params (dict) -- User provided values to replace the placeholders of the prompt template.
metadata (bool, optional) -- False(default), True return resource object with all details.

Returns:

A PromptTemplateSubstitutionResponse object.

Return type:

fill_prompt_template_by_id(template_id, input_params, metadata=False)

Replace the placeholders of the prompt template referenced via template_id with user provided values.

Parameters:

template_id (str) -- The ID of the prompt template.
input_params (dict) -- User provided values to replace the placeholders of the prompt template.
metadata (bool, optional) -- False(default), True return resource object with all details.

Returns:

A PromptTemplateSubstitutionResponse object.

Return type:

get_prompt_template_by_id(template_id)

Retrieve a specific version of the prompt template by ID.

Parameters:: template_id (str) -- The ID of the prompt template to retrieve.
Returns:: A PromptTemplateGetResponse object.
Return type:: PromptTemplateGetResponse

get_prompt_template_history(scenario, name, version)

Retrieve the history of edits to the prompt template. Only for imperative managed prompt templates.

Parameters:

scenario (str) -- The scenario name of the prompt template.
name (str) -- The name of the prompt template.
version (str) -- The version ID of the prompt template.

Returns:

A PromptTemplateListResponse object.

Return type:

get_prompt_templates(scenario, name, version, retrieve=None, include_spec=None)

Retrieve the latest version of every prompt template based on the filters.

Parameters:

scenario (str) -- the scenario name of the prompt template.
name (str) -- the name of the prompt template.
version (str) -- the version of the prompt template.
retrieve (str, optional) -- both(default), imperative, declarative
include_spec (bool, optional) -- false(default), true

Returns:

A PromptTemplateListResponse object.

Return type: