openaivec.prompt

This module provides a builder for creating few‑shot prompts, which are used to train large language models (LLMs) by providing them with examples of input/output pairs. The builder allows for the construction of a prompt in a structured way, including setting the purpose, adding cautions, and providing examples.

from openaivec.prompt import FewShotPromptBuilder

prompt_str: str = (
    FewShotPromptBuilder()
    .purpose("some purpose")
    .caution("some caution")
    .caution("some other caution")
    .example("some input", "some output")
    .example("some other input", "some other output")
    .build()
)
print(prompt_str)

this will produce an XML string that looks like this:

<Prompt>
    <Purpose>some purpose</Purpose>
    <Cautions>
        <Caution>some caution</Caution>
        <Caution>some other caution</Caution>
    </Cautions>
    <Examples>
        <Example>
            <Input>some input</Input>
            <Output>some output</Output>
        </Example>
        <Example>
            <Input>some other input</Input>
            <Output>some other output</Output>
        </Example>
    </Examples>
</Prompt>

Example

Bases: BaseModel

Represents a single input/output example used in few‑shot prompts.

Attributes:

Name	Type	Description
input	str	The input text that will be passed to the model.
output	str	The expected output corresponding to the given input.

Source code in src/openaivec/prompt.py

class Example(BaseModel):
    """Represents a single input/output example used in few‑shot prompts.

    Attributes:
        input (str): The input text that will be passed to the model.
        output (str): The expected output corresponding to the given input.
    """

    input: str
    output: str

FewShotPrompt

Bases: BaseModel

Represents a prompt definition used for few‑shot learning.

The data collected in this model is later rendered into XML and sent to a large‑language model as part of the system prompt.

Attributes:

Name	Type	Description
purpose	str	A concise, human‑readable statement describing the goal of the prompt.
cautions	list[str]	A list of warnings, edge cases, or pitfalls that the model should be aware of when generating answers.
examples	list[Example]	Input/output pairs demonstrating the expected behaviour for a variety of scenarios.

Source code in src/openaivec/prompt.py

class FewShotPrompt(BaseModel):
    """Represents a prompt definition used for few‑shot learning.

    The data collected in this model is later rendered into XML and sent to a
    large‑language model as part of the system prompt.

    Attributes:
        purpose (str): A concise, human‑readable statement describing the goal
            of the prompt.
        cautions (list[str]): A list of warnings, edge cases, or pitfalls that
            the model should be aware of when generating answers.
        examples (list[Example]): Input/output pairs demonstrating the expected
            behaviour for a variety of scenarios.
    """

    purpose: str
    cautions: List[str]
    examples: List[Example]

FewShotPromptBuilder

Source code in src/openaivec/prompt.py

class FewShotPromptBuilder:
    _prompt: FewShotPrompt
    _steps: List[Step]

    def __init__(self):
        """Initialize an empty FewShotPromptBuilder."""
        self._prompt = FewShotPrompt(purpose="", cautions=[], examples=[])

    @classmethod
    def of(cls, prompt: FewShotPrompt) -> "FewShotPromptBuilder":
        """Create a builder pre‑populated with an existing prompt.

        Args:
            prompt (FewShotPrompt): The prompt to start from.

        Returns:
            FewShotPromptBuilder: A new builder instance.
        """
        builder = cls()
        builder._prompt = prompt
        return builder

    @classmethod
    def of_empty(cls) -> "FewShotPromptBuilder":
        """Create a builder.

        Returns:
            FewShotPromptBuilder: A new builder instance with an empty prompt.
        """
        return cls.of(FewShotPrompt(purpose="", cautions=[], examples=[]))

    def purpose(self, purpose: str) -> "FewShotPromptBuilder":
        """Set the purpose of the prompt.

        Args:
            purpose (str): A concise statement describing the prompt’s goal.

        Returns:
            FewShotPromptBuilder: The current builder instance (for chaining).
        """
        self._prompt.purpose = purpose
        return self

    def caution(self, caution: str) -> "FewShotPromptBuilder":
        """Append a cautionary note to the prompt.

        Args:
            caution (str): A caution or edge‑case description.

        Returns:
            FewShotPromptBuilder: The current builder instance.
        """
        if self._prompt.cautions is None:
            self._prompt.cautions = []
        self._prompt.cautions.append(caution)
        return self

    def example(
        self,
        input_value: str | BaseModel,
        output_value: str | BaseModel,
    ) -> "FewShotPromptBuilder":
        """Add a single input/output example.

        Args:
            input_value (str | BaseModel): Example input; if a Pydantic model is
                provided it is serialised to JSON.
            output_value (str | BaseModel): Expected output; serialised if needed.

        Returns:
            FewShotPromptBuilder: The current builder instance.
        """
        if self._prompt.examples is None:
            self._prompt.examples = []

        input_string = input_value if isinstance(input_value, str) else input_value.model_dump_json()
        output_string = output_value if isinstance(output_value, str) else output_value.model_dump_json()
        self._prompt.examples.append(Example(input=input_string, output=output_string))
        return self

    def improve(
        self,
        client: OpenAI,
        model_name: str,
        temperature: float = 0.0,
        top_p: float = 1.0,
    ) -> "FewShotPromptBuilder":
        """Iteratively refine the prompt using an LLM.

        The method calls a single LLM request that returns multiple
        editing steps and stores each step for inspection.

        Args:
            client (openai.OpenAI): Configured OpenAI client.
            model_name (str): Model identifier (e.g. ``gpt-4o-mini``).
            temperature (float, optional): Sampling temperature. Defaults to 0.0.
            top_p (float, optional): Nucleus sampling parameter. Defaults to 1.0.

        Returns:
            FewShotPromptBuilder: The current builder instance containing the refined prompt and iteration history.
        """

        response: ParsedResponse[Response] = client.responses.parse(
            model=model_name,
            instructions=_PROMPT,
            input=Request(prompt=self._prompt).model_dump_json(),
            temperature=temperature,
            top_p=top_p,
            text_format=Response,
        )

        # keep the original prompt
        self._steps = [Step(id=0, analysis="Original Prompt", prompt=self._prompt)]

        # add the histories
        for step in response.output_parsed.iterations:
            self._steps.append(step)

        # set the final prompt
        self._prompt = self._steps[-1].prompt

        return self

    def explain(self) -> "FewShotPromptBuilder":
        """Pretty‑print the diff of each improvement iteration.

        Returns:
            FewShotPromptBuilder: The current builder instance.
        """
        for previous, current in zip(self._steps, self._steps[1:]):
            print(f"=== Iteration {current.id} ===\n")
            print(f"Instruction: {current.analysis}")
            diff = difflib.unified_diff(
                _render_prompt(previous.prompt).splitlines(),
                _render_prompt(current.prompt).splitlines(),
                fromfile="before",
                tofile="after",
                lineterm="",
            )
            for line in diff:
                print(line)
        return self

    def _validate(self) -> None:
        """Validate the internal FewShotPrompt.

        Raises:
            ValueError: If required fields such as purpose or examples are
                missing.
        """
        # Validate that 'purpose' and 'examples' are not empty.
        if not self._prompt.purpose:
            raise ValueError("Purpose is required.")
        if not self._prompt.examples or len(self._prompt.examples) == 0:
            raise ValueError("At least one example is required.")

    def get_object(self) -> FewShotPrompt:
        """Return the underlying FewShotPrompt object.

        Returns:
            FewShotPrompt: The validated prompt object.
        """
        self._validate()
        return self._prompt

    def build(self) -> str:
        """Build and return the prompt as XML.

        Returns:
            str: XML representation of the prompt.
        """
        self._validate()
        return self.build_xml()

    def build_json(self, **kwargs) -> str:
        """Build and return the prompt as a JSON string.

        Args:
            **kwargs: Keyword arguments forwarded to ``model_dump_json``.

        Returns:
            str: JSON representation of the prompt.
        """
        self._validate()
        return self._prompt.model_dump_json(**kwargs)

    def build_xml(self) -> str:
        """Alias for :py:meth:`build` for explicit XML generation.

        Returns:
            str: XML representation of the prompt.
        """
        self._validate()
        return _render_prompt(self._prompt)

__init__()

Initialize an empty FewShotPromptBuilder.

Source code in src/openaivec/prompt.py

def __init__(self):
    """Initialize an empty FewShotPromptBuilder."""
    self._prompt = FewShotPrompt(purpose="", cautions=[], examples=[])

build()

Build and return the prompt as XML.

Returns:

Name	Type	Description
str	str	XML representation of the prompt.

Source code in src/openaivec/prompt.py

def build(self) -> str:
    """Build and return the prompt as XML.

    Returns:
        str: XML representation of the prompt.
    """
    self._validate()
    return self.build_xml()

build_json(**kwargs)

Build and return the prompt as a JSON string.

Parameters:

Name	Type	Description	Default
**kwargs		Keyword arguments forwarded to model_dump_json.	{}

Returns:

Name	Type	Description
str	str	JSON representation of the prompt.

Source code in src/openaivec/prompt.py

def build_json(self, **kwargs) -> str:
    """Build and return the prompt as a JSON string.

    Args:
        **kwargs: Keyword arguments forwarded to ``model_dump_json``.

    Returns:
        str: JSON representation of the prompt.
    """
    self._validate()
    return self._prompt.model_dump_json(**kwargs)

build_xml()

Alias for :py:meth:build for explicit XML generation.

Returns:

Name	Type	Description
str	str	XML representation of the prompt.

Source code in src/openaivec/prompt.py

def build_xml(self) -> str:
    """Alias for :py:meth:`build` for explicit XML generation.

    Returns:
        str: XML representation of the prompt.
    """
    self._validate()
    return _render_prompt(self._prompt)

caution(caution)

Append a cautionary note to the prompt.

Parameters:

Name	Type	Description	Default
caution	str	A caution or edge‑case description.	required

Returns:

Name	Type	Description
FewShotPromptBuilder	FewShotPromptBuilder	The current builder instance.

Source code in src/openaivec/prompt.py

def caution(self, caution: str) -> "FewShotPromptBuilder":
    """Append a cautionary note to the prompt.

    Args:
        caution (str): A caution or edge‑case description.

    Returns:
        FewShotPromptBuilder: The current builder instance.
    """
    if self._prompt.cautions is None:
        self._prompt.cautions = []
    self._prompt.cautions.append(caution)
    return self

example(input_value, output_value)

Add a single input/output example.

Parameters:

Name	Type	Description	Default
input_value	str | BaseModel	Example input; if a Pydantic model is provided it is serialised to JSON.	required
output_value	str | BaseModel	Expected output; serialised if needed.	required

Returns:

Name	Type	Description
FewShotPromptBuilder	FewShotPromptBuilder	The current builder instance.

Source code in src/openaivec/prompt.py

def example(
    self,
    input_value: str | BaseModel,
    output_value: str | BaseModel,
) -> "FewShotPromptBuilder":
    """Add a single input/output example.

    Args:
        input_value (str | BaseModel): Example input; if a Pydantic model is
            provided it is serialised to JSON.
        output_value (str | BaseModel): Expected output; serialised if needed.

    Returns:
        FewShotPromptBuilder: The current builder instance.
    """
    if self._prompt.examples is None:
        self._prompt.examples = []

    input_string = input_value if isinstance(input_value, str) else input_value.model_dump_json()
    output_string = output_value if isinstance(output_value, str) else output_value.model_dump_json()
    self._prompt.examples.append(Example(input=input_string, output=output_string))
    return self

explain()

Pretty‑print the diff of each improvement iteration.

Returns:

Name	Type	Description
FewShotPromptBuilder	FewShotPromptBuilder	The current builder instance.

Source code in src/openaivec/prompt.py

def explain(self) -> "FewShotPromptBuilder":
    """Pretty‑print the diff of each improvement iteration.

    Returns:
        FewShotPromptBuilder: The current builder instance.
    """
    for previous, current in zip(self._steps, self._steps[1:]):
        print(f"=== Iteration {current.id} ===\n")
        print(f"Instruction: {current.analysis}")
        diff = difflib.unified_diff(
            _render_prompt(previous.prompt).splitlines(),
            _render_prompt(current.prompt).splitlines(),
            fromfile="before",
            tofile="after",
            lineterm="",
        )
        for line in diff:
            print(line)
    return self

get_object()

Return the underlying FewShotPrompt object.

Returns:

Name	Type	Description
FewShotPrompt	FewShotPrompt	The validated prompt object.

Source code in src/openaivec/prompt.py

def get_object(self) -> FewShotPrompt:
    """Return the underlying FewShotPrompt object.

    Returns:
        FewShotPrompt: The validated prompt object.
    """
    self._validate()
    return self._prompt

improve(client, model_name, temperature=0.0, top_p=1.0)

Iteratively refine the prompt using an LLM.

The method calls a single LLM request that returns multiple editing steps and stores each step for inspection.

Parameters:

Name	Type	Description	Default
client	OpenAI	Configured OpenAI client.	required
model_name	str	Model identifier (e.g. gpt-4o-mini).	required
temperature	float	Sampling temperature. Defaults to 0.0.	0.0
top_p	float	Nucleus sampling parameter. Defaults to 1.0.	1.0

Returns:

Name	Type	Description
FewShotPromptBuilder	FewShotPromptBuilder	The current builder instance containing the refined prompt and iteration history.

Source code in src/openaivec/prompt.py

def improve(
    self,
    client: OpenAI,
    model_name: str,
    temperature: float = 0.0,
    top_p: float = 1.0,
) -> "FewShotPromptBuilder":
    """Iteratively refine the prompt using an LLM.

    The method calls a single LLM request that returns multiple
    editing steps and stores each step for inspection.

    Args:
        client (openai.OpenAI): Configured OpenAI client.
        model_name (str): Model identifier (e.g. ``gpt-4o-mini``).
        temperature (float, optional): Sampling temperature. Defaults to 0.0.
        top_p (float, optional): Nucleus sampling parameter. Defaults to 1.0.

    Returns:
        FewShotPromptBuilder: The current builder instance containing the refined prompt and iteration history.
    """

    response: ParsedResponse[Response] = client.responses.parse(
        model=model_name,
        instructions=_PROMPT,
        input=Request(prompt=self._prompt).model_dump_json(),
        temperature=temperature,
        top_p=top_p,
        text_format=Response,
    )

    # keep the original prompt
    self._steps = [Step(id=0, analysis="Original Prompt", prompt=self._prompt)]

    # add the histories
    for step in response.output_parsed.iterations:
        self._steps.append(step)

    # set the final prompt
    self._prompt = self._steps[-1].prompt

    return self

of(prompt) classmethod

Create a builder pre‑populated with an existing prompt.

Parameters:

Name	Type	Description	Default
prompt	FewShotPrompt	The prompt to start from.	required

Returns:

Name	Type	Description
FewShotPromptBuilder	FewShotPromptBuilder	A new builder instance.

Source code in src/openaivec/prompt.py

@classmethod
def of(cls, prompt: FewShotPrompt) -> "FewShotPromptBuilder":
    """Create a builder pre‑populated with an existing prompt.

    Args:
        prompt (FewShotPrompt): The prompt to start from.

    Returns:
        FewShotPromptBuilder: A new builder instance.
    """
    builder = cls()
    builder._prompt = prompt
    return builder

of_empty() classmethod

Create a builder.

Returns:

Name	Type	Description
FewShotPromptBuilder	FewShotPromptBuilder	A new builder instance with an empty prompt.

Source code in src/openaivec/prompt.py

@classmethod
def of_empty(cls) -> "FewShotPromptBuilder":
    """Create a builder.

    Returns:
        FewShotPromptBuilder: A new builder instance with an empty prompt.
    """
    return cls.of(FewShotPrompt(purpose="", cautions=[], examples=[]))

purpose(purpose)

Set the purpose of the prompt.

Parameters:

Name	Type	Description	Default
purpose	str	A concise statement describing the prompt’s goal.	required

Returns:

Name	Type	Description
FewShotPromptBuilder	FewShotPromptBuilder	The current builder instance (for chaining).

Source code in src/openaivec/prompt.py

def purpose(self, purpose: str) -> "FewShotPromptBuilder":
    """Set the purpose of the prompt.

    Args:
        purpose (str): A concise statement describing the prompt’s goal.

    Returns:
        FewShotPromptBuilder: The current builder instance (for chaining).
    """
    self._prompt.purpose = purpose
    return self

Step

Bases: BaseModel

A single refinement iteration produced by the LLM.

Attributes:

Name	Type	Description
id	int	Sequential identifier of the iteration (0 for the original, 1 for the first change, and so on).
analysis	str	Natural‑language explanation of the issue addressed in this iteration and why the change was necessary.
prompt	FewShotPrompt	The updated prompt after applying the described modification.

Source code in src/openaivec/prompt.py

class Step(BaseModel):
    """A single refinement iteration produced by the LLM.

    Attributes:
        id (int): Sequential identifier of the iteration (``0`` for the
            original, ``1`` for the first change, and so on).
        analysis (str): Natural‑language explanation of the issue addressed
            in this iteration and why the change was necessary.
        prompt (FewShotPrompt): The updated prompt after applying the
            described modification.
    """

    id: int
    analysis: str
    prompt: FewShotPrompt