Catalog

Modules concerning the catalog to manage and load prompt builders used in Gen AI applications.

LMRequestProcessorCatalog

Bases: BaseCatalog[LMRequestProcessor]

Loads multiple LM request processors from certain sources.

Attributes:

Name	Type	Description
components	dict[str, LMRequestProcessor]	Dictionary of the loaded LM request processors.

Initialization

Example 1: Load from Google Sheets using client email and private key

catalog = LMRequestProcessorCatalog.from_gsheets(
    sheet_id="...",
    worksheet_id="...",
    client_email="...",
    private_key="...",
)

lm_request_processor = catalog.name

Example 2: Load from Google Sheets using credential file

catalog = LMRequestProcessorCatalog.from_gsheets(
    sheet_id="...",
    worksheet_id="...",
    credential_file_path="...",
)

lm_request_processor = catalog.name

Example 3: Load from CSV

catalog = LMRequestProcessorCatalog.from_csv(csv_path="...")

lm_request_processor = catalog.name

Example 4: Load from record

catalog = LMRequestProcessorCatalog.from_records(
    name="...",
    system_template="...",
    user_template="...",
    model_id="...",
    credentials="...",
    config="...",
    output_parser_type="...",
)

lm_request_processor = catalog.name

Template Format Example

Example 1: Google Sheets

For an example of how a Google Sheets file can be formatted to be loaded using LMRequestProcessorCatalog, see: https://docs.google.com/spreadsheets/d/1CX9i45yEinv1UdB3s6uHNMj7mxr2-s1NFHfFDvMsq0E/edit?usp=drive_link

Example 2: CSV

For an example of how a CSV file can be formatted to be loaded using LMRequestProcessorCatalog, see: https://drive.google.com/file/d/10nYKn_r9SVnTkaik-caMqUjX6prUZ62M/view?usp=drive_link

Template Explanation

The required columns are: 1. name (str): The name of the LM request processor. 2. system_template (str): The system template of the prompt builder. 3. user_template (str): The user template of the prompt builder. 4. model_id (str): The model ID of the LM invoker. 5. credentials (str | json_str): The credentials of the LM invoker. 6. config (json_str): The additional configuration of the LM invoker. 7. output_parser_type (str): The type of the output parser.

Important Notes: 1. At least one of system_template or user_template must be filled. 2. The model_id: 2.1. Must be filled with the model ID of the LM invoker, e.g. "openai/gpt-4.1-nano". 2.2. Can be partially loaded from the environment variable using the "${ENV_VAR_KEY}" syntax, e.g. "azure-openai/${AZURE_ENDPOINT}/${AZURE_DEPLOYMENT}". 2.3. For the available model ID formats, see: https://gdplabs.gitbook.io/sdk/resources/supported-models 3. credentials is optional. If it is filled, it can either be: 3.1. An environment variable name containing the API key (e.g. OPENAI_API_KEY). 3.2. A path to a credentials JSON file, currently only supported for Google Vertex AI. 3.3. A dictionary of credentials, currently supported for Bedrock and LangChain. 4. config is optional. When this column is empty, the LM invoker will use the default configuration. If it is filled, it must be a valid JSON string. 5. output_parser_type can either be: 5.1. none: No output parser will be used. 5.2. json: The JSONOutputParser will be used.

PromptBuilderCatalog

Bases: BaseCatalog[BasePromptBuilder | MultimodalPromptBuilder | PromptBuilder]

Loads multiple prompt builders from certain sources.

Attributes:

Name	Type	Description
components	dict[str, BasePromptBuilder | MultimodalPromptBuilder | PromptBuilder]	Dictionary of the loaded prompt builders.

Initialization

Example 1: Load from Google Sheets using client email and private key

catalog = PromptBuilderCatalog.from_gsheets(
    sheet_id="...",
    worksheet_id="...",
    client_email="...",
    private_key="...",
)

prompt_builder = catalog.name

Example 2: Load from Google Sheets using credential file

catalog = PromptBuilderCatalog.from_gsheets(
    sheet_id="...",
    worksheet_id="...",
    credential_file_path="...",
)

prompt_builder = catalog.name

Example 3: Load from CSV

catalog = PromptBuilderCatalog.from_csv(csv_path="...")

prompt_builder = catalog.name

Example 4: Load from records

catalog = PromptBuilderCatalog.from_records(
    records=[
        {
            "name": "summarize",
            "system": "You are an AI expert\nSummarize the following context.\n\nContext:\n```{context}```",
            "user": ""
        },
        {
            "name": "transform_query",
            "system": "",
            "user": "Transform the following query into a simpler form.\n\nQuery:\n```{query}```"
        },
        {
            "name": "draft_document",
            "system": (
                "You are an AI expert.\nDraft a document following the provided format and context.\n\n"
                "Format:\n```{format}```\n\nContext:\n```{context}```"
            ),
            "user": "User instruction:\n{query}"
        },
    ]
)

prompt_builder = catalog.answer_question

Template Example

Example 1: Google Sheets

For an example of how a Google Sheets file can be formatted to be loaded using PromptBuilderCatalog, see: https://docs.google.com/spreadsheets/d/12IwSKv8hMhyWXSQnLx9LgCj0cxaR1f9gOmbEDGleurE/edit?usp=drive_link

Example 2: CSV

For an example of how a CSV file can be formatted to be loaded using PromptBuilderCatalog, see: https://drive.google.com/file/d/1CWijOk-g16ZglUn_K2bDPmbyyBDK2r0L/view?usp=drive_link

Template explanation

The required columns are: 1. name (str): The name of the prompt builder. 2. system (str): The system template of the prompt builder. 3. user (str): The user template of the prompt builder.

Important Notes: 1. At least one of the system and user columns must be filled.

WARNING: The use of BasePromptBuilder | MultimodalPromptBuilder is deprecated and will be removed in version 0.5.0. Please use PromptBuilder instead.