Skip to content

Prompt templates

This section documents the prompt template classes.

hf_hub_prompts.prompt_templates

BasePromptTemplate

Bases: ABC

An abstract base class for prompt templates.

This class defines the common interface and shared functionality for all prompt templates. Users should not instantiate this class directly, but instead use TextPromptTemplate or ChatPromptTemplate, which are subclasses of BasePromptTemplate.

Source code in hf_hub_prompts/prompt_templates.py 
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
class BasePromptTemplate(ABC):
    """An abstract base class for prompt templates.

    This class defines the common interface and shared functionality for all prompt templates.
    Users should not instantiate this class directly, but instead use TextPromptTemplate
    or ChatPromptTemplate, which are subclasses of BasePromptTemplate.
    """

    # Type hints for optional standard attributes shared across all template types
    metadata: Optional[Dict[str, Any]]
    input_variables: Optional[List[str]]
    other_data: Dict[str, Any]

    def __init__(self, prompt_data: Dict[str, Any], prompt_url: Optional[str] = None) -> None:
        # Set template-specific required attributes
        self._set_required_attributes_for_template_type(prompt_data)

        # Set optional standard attributes that are the same across all templates
        self.input_variables = prompt_data.get("input_variables")
        self.metadata = prompt_data.get("metadata")

        # Store any additional optional data that might be present in the prompt data
        self.other_data = {
            k: v
            for k, v in prompt_data.items()
            if k not in ["metadata", "input_variables"] + self._get_required_attributes_for_template_type()
        }

        if prompt_url is not None:
            self.other_data["prompt_url"] = prompt_url

    @abstractmethod
    def _get_required_attributes_for_template_type(self) -> List[str]:
        """Return list of required keys for this template type."""
        pass

    @abstractmethod
    def _set_required_attributes_for_template_type(self, prompt_data: Dict[str, Any]) -> None:
        """Set required attributes for this template type."""
        pass

    @abstractmethod
    def populate_template(self, **input_variables: Any) -> PopulatedPrompt:
        """Abstract method to populate the prompt template with the given variables.

        Args:
            **input_variables: The values to fill placeholders in the template.

        Returns:
            PopulatedPrompt: A PopulatedPrompt object containing the populated content.
        """
        pass

    def display(self, format: Literal["json", "yaml"] = "json") -> None:
        """Display the prompt configuration in the specified format.

        Examples:
            >>> from hf_hub_prompts import download_prompt_template
            >>> prompt_template = download_prompt_template(
            ...     repo_id="MoritzLaurer/example_prompts",
            ...     filename="translate.yaml"
            ... )
            >>> prompt_template.display(format="yaml")  # doctest: +NORMALIZE_WHITESPACE
            template: 'Translate the following text to {language}:
              {text}'
            input_variables:
            - language
            - text
            metadata:
              name: Simple Translator
              description: A simple translation prompt for illustrating the standard prompt YAML
                format
              tags:
              - translation
              - multilinguality
              version: 0.0.1
              author: Some Person
        """
        # Create a dict of all attributes except other_data
        display_dict = self.__dict__.copy()
        display_dict.pop("other_data", None)

        if format == "json":
            print(json.dumps(display_dict, indent=2), end="")
        elif format == "yaml":
            print(yaml.dump(display_dict, default_flow_style=False, sort_keys=False), end="")

    def to_dict(self) -> Dict[str, Any]:
        return self.__dict__

    def __getitem__(self, key: str) -> Any:
        return self.__dict__[key]

    def __repr__(self) -> str:
        attributes = ", ".join(
            f"{key}={repr(value)[:50]}..." if len(repr(value)) > 50 else f"{key}={repr(value)}"
            for key, value in self.__dict__.items()
        )
        return f"{self.__class__.__name__}({attributes})"

    def _fill_placeholders(self, template_part: Any, input_variables: Dict[str, Any]) -> Any:
        """Recursively fill placeholders in strings or nested structures like dicts or lists."""
        pattern = re.compile(r"\{([^{}]+)\}")

        if isinstance(template_part, str):
            # fill placeholders in strings
            def replacer(match: Match[str]) -> str:
                key = match.group(1).strip()
                return str(input_variables.get(key, match.group(0)))

            return pattern.sub(replacer, template_part)

        elif isinstance(template_part, dict):
            # Recursively handle dictionaries
            return {key: self._fill_placeholders(value, input_variables) for key, value in template_part.items()}

        elif isinstance(template_part, list):
            # Recursively handle lists
            return [self._fill_placeholders(item, input_variables) for item in template_part]

        return template_part  # For non-string, non-dict, non-list types, return as is

    def _validate_input_variables(self, input_variables: Dict[str, Any]) -> None:
        """Validate that the provided input variables match the expected ones.

        Args:
            input_variables: Dictionary of variables to validate.

        Behavior:
            - If prompt_template.input_variables is defined:
                Ensures exact match between provided and expected variables.
            - If prompt_template.input_variables is not defined:
                Skips validation (logs warning) and accepts any variables.

        Raises:
            ValueError: If prompt_template.input_variables is defined and there are
                missing or unexpected variables.
        """
        if self.input_variables:
            missing_vars = set(self.input_variables) - set(input_variables.keys())
            extra_vars = set(input_variables.keys()) - set(self.input_variables)

            if missing_vars or extra_vars:
                error_msg = []
                error_msg.append(f"Expected input_variables from the prompt template: {self.input_variables}")
                if missing_vars:
                    error_msg.append(f"Missing variables: {list(missing_vars)}")
                if extra_vars:
                    error_msg.append(f"Unexpected variables: {list(extra_vars)}")
                if self.other_data["prompt_url"]:
                    error_msg.append(f"Template URL: {self.other_data["prompt_url"]}")

                raise ValueError("\n".join(error_msg))
        else:
            logger.warning(
                "No input_variables specified in prompt template. Input validation is disabled. "
                "To enable validation, specify input_variables when creating the prompt template. "
                "Without validation, misspelled variable names will be left unreplaced in the output."
            )

populate_template abstractmethod 

populate_template(**input_variables)

Abstract method to populate the prompt template with the given variables.

Parameters:

Name Type Description Default
**input_variables Any

The values to fill placeholders in the template.

 {}

Returns:

Name Type Description
PopulatedPrompt PopulatedPrompt

A PopulatedPrompt object containing the populated content.
Source code in hf_hub_prompts/prompt_templates.py 
64
65
66
67
68
69
70
71
72
73
74
@abstractmethod
def populate_template(self, **input_variables: Any) -> PopulatedPrompt:
    """Abstract method to populate the prompt template with the given variables.

    Args:
        **input_variables: The values to fill placeholders in the template.

    Returns:
        PopulatedPrompt: A PopulatedPrompt object containing the populated content.
    """
    pass

display 

display(format='json')

Display the prompt configuration in the specified format.

Examples:

>>> from hf_hub_prompts import download_prompt_template
>>> prompt_template = download_prompt_template(
...     repo_id="MoritzLaurer/example_prompts",
...     filename="translate.yaml"
... )
>>> prompt_template.display(format="yaml")
template: 'Translate the following text to {language}:
  {text}'
input_variables:
- language
- text
metadata:
  name: Simple Translator
  description: A simple translation prompt for illustrating the standard prompt YAML
    format
  tags:
  - translation
  - multilinguality
  version: 0.0.1
  author: Some Person
Source code in hf_hub_prompts/prompt_templates.py 
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
def display(self, format: Literal["json", "yaml"] = "json") -> None:
    """Display the prompt configuration in the specified format.

    Examples:
        >>> from hf_hub_prompts import download_prompt_template
        >>> prompt_template = download_prompt_template(
        ...     repo_id="MoritzLaurer/example_prompts",
        ...     filename="translate.yaml"
        ... )
        >>> prompt_template.display(format="yaml")  # doctest: +NORMALIZE_WHITESPACE
        template: 'Translate the following text to {language}:
          {text}'
        input_variables:
        - language
        - text
        metadata:
          name: Simple Translator
          description: A simple translation prompt for illustrating the standard prompt YAML
            format
          tags:
          - translation
          - multilinguality
          version: 0.0.1
          author: Some Person
    """
    # Create a dict of all attributes except other_data
    display_dict = self.__dict__.copy()
    display_dict.pop("other_data", None)

    if format == "json":
        print(json.dumps(display_dict, indent=2), end="")
    elif format == "yaml":
        print(yaml.dump(display_dict, default_flow_style=False, sort_keys=False), end="")

TextPromptTemplate

Bases: BasePromptTemplate

A class representing a standard text prompt template.

Examples:

Download and use a text prompt template:

>>> from hf_hub_prompts import download_prompt_template
>>> # Download example translation prompt
>>> prompt_template = download_prompt_template(
...     repo_id="MoritzLaurer/example_prompts",
...     filename="translate.yaml"
... )
>>> # Inspect template attributes
>>> prompt_template.template
'Translate the following text to {language}:\n{text}'
>>> prompt_template.input_variables
['language', 'text']
>>> prompt_template.metadata['name']
'Simple Translator'

>>> # Use the template
>>> prompt = prompt_template.populate_template(
...     language="French",
...     text="Hello world!"
... )
>>> prompt.content
'Translate the following text to French:\nHello world!'
Source code in hf_hub_prompts/prompt_templates.py 
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
class TextPromptTemplate(BasePromptTemplate):
    """A class representing a standard text prompt template.

    Examples:
        Download and use a text prompt template:
        >>> from hf_hub_prompts import download_prompt_template
        >>> # Download example translation prompt
        >>> prompt_template = download_prompt_template(
        ...     repo_id="MoritzLaurer/example_prompts",
        ...     filename="translate.yaml"
        ... )
        >>> # Inspect template attributes
        >>> prompt_template.template
        'Translate the following text to {language}:\\n{text}'
        >>> prompt_template.input_variables
        ['language', 'text']
        >>> prompt_template.metadata['name']
        'Simple Translator'

        >>> # Use the template
        >>> prompt = prompt_template.populate_template(
        ...     language="French",
        ...     text="Hello world!"
        ... )
        >>> prompt.content
        'Translate the following text to French:\\nHello world!'
    """

    # Type hints for template-specific attributes
    template: str

    def _get_required_attributes_for_template_type(self) -> List[str]:
        return ["template"]

    def _set_required_attributes_for_template_type(self, prompt_data: Dict[str, Any]) -> None:
        if "template" not in prompt_data:
            raise ValueError("You must provide 'template' in prompt_data")
        self.template = prompt_data["template"]

    def populate_template(self, **input_variables: Any) -> PopulatedPrompt:
        """Populate the prompt by replacing placeholders with provided values.

        Examples:
            >>> from hf_hub_prompts import download_prompt_template
            >>> prompt_template = download_prompt_template(
            ...     repo_id="MoritzLaurer/example_prompts",
            ...     filename="translate.yaml"
            ... )
            >>> prompt_template.template
            'Translate the following text to {language}:\\n{text}'
            >>> prompt = prompt_template.populate_template(
            ...     language="French",
            ...     text="Hello world!"
            ... )
            >>> prompt.content
            'Translate the following text to French:\\nHello world!'

        Args:
            **input_variables: The values to fill placeholders in the prompt template.

        Returns:
            PopulatedPrompt: A PopulatedPrompt object containing the populated prompt string.
        """
        self._validate_input_variables(input_variables)
        populated_prompt = self._fill_placeholders(self.template, input_variables)
        return PopulatedPrompt(content=populated_prompt)

    def to_langchain_template(self) -> "LC_PromptTemplate":
        """Convert the TextPromptTemplate to a LangChain PromptTemplate.

        Examples:
            >>> from hf_hub_prompts import download_prompt_template
            >>> prompt_template = download_prompt_template(
            ...     repo_id="MoritzLaurer/example_prompts",
            ...     filename="translate.yaml"
            ... )
            >>> lc_template = prompt_template.to_langchain_template()
            >>> # test equivalence
            >>> from langchain_core.prompts import PromptTemplate as LC_PromptTemplate
            >>> isinstance(lc_template, LC_PromptTemplate)
            True

        Returns:
            PromptTemplate: A LangChain PromptTemplate object.

        Raises:
            ImportError: If LangChain is not installed.
        """
        try:
            from langchain_core.prompts import PromptTemplate as LC_PromptTemplate
        except ImportError as e:
            raise ImportError("LangChain is not installed. Please install it with 'pip install langchain'") from e

        return LC_PromptTemplate(
            template=self.template,
            input_variables=self.input_variables,
            metadata=self.metadata,
        )

populate_template 

populate_template(**input_variables)

Populate the prompt by replacing placeholders with provided values.

Examples:

>>> from hf_hub_prompts import download_prompt_template
>>> prompt_template = download_prompt_template(
...     repo_id="MoritzLaurer/example_prompts",
...     filename="translate.yaml"
... )
>>> prompt_template.template
'Translate the following text to {language}:\n{text}'
>>> prompt = prompt_template.populate_template(
...     language="French",
...     text="Hello world!"
... )
>>> prompt.content
'Translate the following text to French:\nHello world!'

Parameters:

Name Type Description Default
**input_variables Any

The values to fill placeholders in the prompt template.

 {}

Returns:

Name Type Description
PopulatedPrompt PopulatedPrompt

A PopulatedPrompt object containing the populated prompt string.
Source code in hf_hub_prompts/prompt_templates.py 
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
def populate_template(self, **input_variables: Any) -> PopulatedPrompt:
    """Populate the prompt by replacing placeholders with provided values.

    Examples:
        >>> from hf_hub_prompts import download_prompt_template
        >>> prompt_template = download_prompt_template(
        ...     repo_id="MoritzLaurer/example_prompts",
        ...     filename="translate.yaml"
        ... )
        >>> prompt_template.template
        'Translate the following text to {language}:\\n{text}'
        >>> prompt = prompt_template.populate_template(
        ...     language="French",
        ...     text="Hello world!"
        ... )
        >>> prompt.content
        'Translate the following text to French:\\nHello world!'

    Args:
        **input_variables: The values to fill placeholders in the prompt template.

    Returns:
        PopulatedPrompt: A PopulatedPrompt object containing the populated prompt string.
    """
    self._validate_input_variables(input_variables)
    populated_prompt = self._fill_placeholders(self.template, input_variables)
    return PopulatedPrompt(content=populated_prompt)

to_langchain_template 

to_langchain_template()

Convert the TextPromptTemplate to a LangChain PromptTemplate.

Examples:

>>> from hf_hub_prompts import download_prompt_template
>>> prompt_template = download_prompt_template(
...     repo_id="MoritzLaurer/example_prompts",
...     filename="translate.yaml"
... )
>>> lc_template = prompt_template.to_langchain_template()
>>> # test equivalence
>>> from langchain_core.prompts import PromptTemplate as LC_PromptTemplate
>>> isinstance(lc_template, LC_PromptTemplate)
True

Returns:

Name Type Description
PromptTemplate PromptTemplate

A LangChain PromptTemplate object.

Raises:

Type Description
ImportError

If LangChain is not installed.
Source code in hf_hub_prompts/prompt_templates.py 
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
def to_langchain_template(self) -> "LC_PromptTemplate":
    """Convert the TextPromptTemplate to a LangChain PromptTemplate.

    Examples:
        >>> from hf_hub_prompts import download_prompt_template
        >>> prompt_template = download_prompt_template(
        ...     repo_id="MoritzLaurer/example_prompts",
        ...     filename="translate.yaml"
        ... )
        >>> lc_template = prompt_template.to_langchain_template()
        >>> # test equivalence
        >>> from langchain_core.prompts import PromptTemplate as LC_PromptTemplate
        >>> isinstance(lc_template, LC_PromptTemplate)
        True

    Returns:
        PromptTemplate: A LangChain PromptTemplate object.

    Raises:
        ImportError: If LangChain is not installed.
    """
    try:
        from langchain_core.prompts import PromptTemplate as LC_PromptTemplate
    except ImportError as e:
        raise ImportError("LangChain is not installed. Please install it with 'pip install langchain'") from e

    return LC_PromptTemplate(
        template=self.template,
        input_variables=self.input_variables,
        metadata=self.metadata,
    )

ChatPromptTemplate

Bases: BasePromptTemplate

A class representing a chat prompt template that can be formatted for and used with various LLM clients.

Examples:

Download and use a chat prompt template:

>>> from hf_hub_prompts import download_prompt_template
>>> # Download example code teaching prompt
>>> prompt_template = download_prompt_template(
...     repo_id="MoritzLaurer/example_prompts",
...     filename="code_teacher.yaml"
... )
>>> # Inspect template attributes
>>> prompt_template.messages
[{'role': 'system', 'content': 'You are a coding assistant who explains concepts clearly and provides short examples.'}, {'role': 'user', 'content': 'Explain what {concept} is in {programming_language}.'}]
>>> prompt_template.input_variables
['concept', 'programming_language']

>>> # Populate the template
>>> prompt = prompt_template.populate_template(
...     concept="list comprehension",
...     programming_language="Python"
... )
>>> prompt.content
[{'role': 'system', 'content': 'You are a coding assistant who explains concepts clearly and provides short examples.'}, {'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]

>>> # By default, the populated prompt is in the OpenAI messages format, as it is adopted by many open-source libraries
>>> # You can convert to formats used by other LLM clients like Anthropic like this:
>>> messages_anthropic = prompt.format_for_client("anthropic")
>>> messages_anthropic
{'system': 'You are a coding assistant who explains concepts clearly and provides short examples.', 'messages': [{'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]}

>>> # Convenience method to populate and format in one step
>>> messages = prompt_template.create_messages(
...     client="anthropic",
...     concept="list comprehension",
...     programming_language="Python"
... )
>>> messages
{'system': 'You are a coding assistant who explains concepts clearly and provides short examples.', 'messages': [{'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]}
Source code in hf_hub_prompts/prompt_templates.py 
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
class ChatPromptTemplate(BasePromptTemplate):
    """A class representing a chat prompt template that can be formatted for and used with various LLM clients.

    Examples:
        Download and use a chat prompt template:
        >>> from hf_hub_prompts import download_prompt_template
        >>> # Download example code teaching prompt
        >>> prompt_template = download_prompt_template(
        ...     repo_id="MoritzLaurer/example_prompts",
        ...     filename="code_teacher.yaml"
        ... )
        >>> # Inspect template attributes
        >>> prompt_template.messages
        [{'role': 'system', 'content': 'You are a coding assistant who explains concepts clearly and provides short examples.'}, {'role': 'user', 'content': 'Explain what {concept} is in {programming_language}.'}]
        >>> prompt_template.input_variables
        ['concept', 'programming_language']

        >>> # Populate the template
        >>> prompt = prompt_template.populate_template(
        ...     concept="list comprehension",
        ...     programming_language="Python"
        ... )
        >>> prompt.content
        [{'role': 'system', 'content': 'You are a coding assistant who explains concepts clearly and provides short examples.'}, {'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]

        >>> # By default, the populated prompt is in the OpenAI messages format, as it is adopted by many open-source libraries
        >>> # You can convert to formats used by other LLM clients like Anthropic like this:
        >>> messages_anthropic = prompt.format_for_client("anthropic")
        >>> messages_anthropic
        {'system': 'You are a coding assistant who explains concepts clearly and provides short examples.', 'messages': [{'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]}

        >>> # Convenience method to populate and format in one step
        >>> messages = prompt_template.create_messages(
        ...     client="anthropic",
        ...     concept="list comprehension",
        ...     programming_language="Python"
        ... )
        >>> messages
        {'system': 'You are a coding assistant who explains concepts clearly and provides short examples.', 'messages': [{'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]}
    """

    # Type hints for template-specific attributes
    messages: List[Dict[str, Any]]

    def _get_required_attributes_for_template_type(self) -> List[str]:
        return ["messages"]

    def _set_required_attributes_for_template_type(self, prompt_data: Dict[str, Any]) -> None:
        if "messages" not in prompt_data:
            raise ValueError("You must provide 'messages' in prompt_data")
        self.messages = prompt_data["messages"]

    def populate_template(self, **input_variables: Any) -> PopulatedPrompt:
        """Populate the prompt messages by replacing placeholders with provided values.

        Examples:
            >>> from hf_hub_prompts import download_prompt_template
            >>> prompt_template = download_prompt_template(
            ...     repo_id="MoritzLaurer/example_prompts",
            ...     filename="code_teacher.yaml"
            ... )
            >>> prompt = prompt_template.populate_template(
            ...     concept="list comprehension",
            ...     programming_language="Python"
            ... )
            >>> prompt.content
            [{'role': 'system', 'content': 'You are a coding assistant who explains concepts clearly and provides short examples.'}, {'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]

        Args:
            **input_variables: The values to fill placeholders in the messages.

        Returns:
            PopulatedPrompt: A PopulatedPrompt object containing the populated messages.
        """
        self._validate_input_variables(input_variables)

        messages_populated = [
            {**msg, "content": self._fill_placeholders(msg["content"], input_variables)} for msg in self.messages
        ]
        return PopulatedPrompt(content=messages_populated)

    def create_messages(
        self, client: str = "openai", **input_variables: Any
    ) -> Union[List[Dict[str, Any]], Dict[str, Any]]:
        """Convenience method to populate a prompt template and format for client in one step.

        Examples:
            >>> from hf_hub_prompts import download_prompt_template
            >>> prompt_template = download_prompt_template(
            ...     repo_id="MoritzLaurer/example_prompts",
            ...     filename="code_teacher.yaml"
            ... )
            >>> # Format for OpenAI (default)
            >>> messages = prompt_template.create_messages(
            ...     concept="list comprehension",
            ...     programming_language="Python"
            ... )
            >>> messages
            [{'role': 'system', 'content': 'You are a coding assistant who explains concepts clearly and provides short examples.'}, {'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]

            >>> # Format for Anthropic
            >>> messages = prompt_template.create_messages(
            ...     client="anthropic",
            ...     concept="list comprehension",
            ...     programming_language="Python"
            ... )
            >>> messages
            {'system': 'You are a coding assistant who explains concepts clearly and provides short examples.', 'messages': [{'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]}

        Args:
            client (str): The client format to use ('openai', 'anthropic'). Defaults to 'openai'.
            **input_variables: The variables to fill into the prompt template. For example, if your template
                expects variables like 'name' and 'age', pass them as keyword arguments:

        Returns:
            Union[List[Dict[str, Any]], Dict[str, Any]]: Populated and formatted messages.
        """
        if "client" in input_variables:
            logger.warning(
                f"'client' was passed both as a parameter for the LLM inference client ('{client}') and in input_variables "
                f"('{input_variables['client']}'). The first parameter version will be used for formatting, "
                "while the second input_variables version will be used in template population."
            )

        prompt = self.populate_template(**input_variables)
        return prompt.format_for_client(client)

    def to_langchain_template(self) -> "LC_ChatPromptTemplate":
        """Convert the ChatPromptTemplate to a LangChain ChatPromptTemplate.

        Examples:
            >>> from hf_hub_prompts import download_prompt_template
            >>> prompt_template = download_prompt_template(
            ...     repo_id="MoritzLaurer/example_prompts",
            ...     filename="code_teacher.yaml"
            ... )
            >>> lc_template = prompt_template.to_langchain_template()
            >>> # test equivalence
            >>> from langchain_core.prompts import ChatPromptTemplate as LC_ChatPromptTemplate
            >>> isinstance(lc_template, LC_ChatPromptTemplate)
            True

        Returns:
            ChatPromptTemplate: A LangChain ChatPromptTemplate object.

        Raises:
            ImportError: If LangChain is not installed.
        """
        try:
            from langchain_core.prompts import ChatPromptTemplate as LC_ChatPromptTemplate
        except ImportError as e:
            raise ImportError("LangChain is not installed. Please install it with 'pip install langchain'") from e

        return LC_ChatPromptTemplate(
            messages=[(msg["role"], msg["content"]) for msg in self.messages],
            input_variables=self.input_variables,
            metadata=self.metadata,
        )

populate_template 

populate_template(**input_variables)

Populate the prompt messages by replacing placeholders with provided values.

Examples:

>>> from hf_hub_prompts import download_prompt_template
>>> prompt_template = download_prompt_template(
...     repo_id="MoritzLaurer/example_prompts",
...     filename="code_teacher.yaml"
... )
>>> prompt = prompt_template.populate_template(
...     concept="list comprehension",
...     programming_language="Python"
... )
>>> prompt.content
[{'role': 'system', 'content': 'You are a coding assistant who explains concepts clearly and provides short examples.'}, {'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]

Parameters:

Name Type Description Default
**input_variables Any

The values to fill placeholders in the messages.

 {}

Returns:

Name Type Description
PopulatedPrompt PopulatedPrompt

A PopulatedPrompt object containing the populated messages.
Source code in hf_hub_prompts/prompt_templates.py 
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
def populate_template(self, **input_variables: Any) -> PopulatedPrompt:
    """Populate the prompt messages by replacing placeholders with provided values.

    Examples:
        >>> from hf_hub_prompts import download_prompt_template
        >>> prompt_template = download_prompt_template(
        ...     repo_id="MoritzLaurer/example_prompts",
        ...     filename="code_teacher.yaml"
        ... )
        >>> prompt = prompt_template.populate_template(
        ...     concept="list comprehension",
        ...     programming_language="Python"
        ... )
        >>> prompt.content
        [{'role': 'system', 'content': 'You are a coding assistant who explains concepts clearly and provides short examples.'}, {'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]

    Args:
        **input_variables: The values to fill placeholders in the messages.

    Returns:
        PopulatedPrompt: A PopulatedPrompt object containing the populated messages.
    """
    self._validate_input_variables(input_variables)

    messages_populated = [
        {**msg, "content": self._fill_placeholders(msg["content"], input_variables)} for msg in self.messages
    ]
    return PopulatedPrompt(content=messages_populated)

create_messages 

create_messages(client='openai', **input_variables)

Convenience method to populate a prompt template and format for client in one step.

Examples:

>>> from hf_hub_prompts import download_prompt_template
>>> prompt_template = download_prompt_template(
...     repo_id="MoritzLaurer/example_prompts",
...     filename="code_teacher.yaml"
... )
>>> # Format for OpenAI (default)
>>> messages = prompt_template.create_messages(
...     concept="list comprehension",
...     programming_language="Python"
... )
>>> messages
[{'role': 'system', 'content': 'You are a coding assistant who explains concepts clearly and provides short examples.'}, {'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]

>>> # Format for Anthropic
>>> messages = prompt_template.create_messages(
...     client="anthropic",
...     concept="list comprehension",
...     programming_language="Python"
... )
>>> messages
{'system': 'You are a coding assistant who explains concepts clearly and provides short examples.', 'messages': [{'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]}

Parameters:

Name Type Description Default
client str

The client format to use ('openai', 'anthropic'). Defaults to 'openai'.

 'openai'
**input_variables Any

The variables to fill into the prompt template. For example, if your template expects variables like 'name' and 'age', pass them as keyword arguments:

 {}

Returns:

Type Description
Union[List[Dict[str, Any]], Dict[str, Any]]

Union[List[Dict[str, Any]], Dict[str, Any]]: Populated and formatted messages.
Source code in hf_hub_prompts/prompt_templates.py 
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
def create_messages(
    self, client: str = "openai", **input_variables: Any
) -> Union[List[Dict[str, Any]], Dict[str, Any]]:
    """Convenience method to populate a prompt template and format for client in one step.

    Examples:
        >>> from hf_hub_prompts import download_prompt_template
        >>> prompt_template = download_prompt_template(
        ...     repo_id="MoritzLaurer/example_prompts",
        ...     filename="code_teacher.yaml"
        ... )
        >>> # Format for OpenAI (default)
        >>> messages = prompt_template.create_messages(
        ...     concept="list comprehension",
        ...     programming_language="Python"
        ... )
        >>> messages
        [{'role': 'system', 'content': 'You are a coding assistant who explains concepts clearly and provides short examples.'}, {'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]

        >>> # Format for Anthropic
        >>> messages = prompt_template.create_messages(
        ...     client="anthropic",
        ...     concept="list comprehension",
        ...     programming_language="Python"
        ... )
        >>> messages
        {'system': 'You are a coding assistant who explains concepts clearly and provides short examples.', 'messages': [{'role': 'user', 'content': 'Explain what list comprehension is in Python.'}]}

    Args:
        client (str): The client format to use ('openai', 'anthropic'). Defaults to 'openai'.
        **input_variables: The variables to fill into the prompt template. For example, if your template
            expects variables like 'name' and 'age', pass them as keyword arguments:

    Returns:
        Union[List[Dict[str, Any]], Dict[str, Any]]: Populated and formatted messages.
    """
    if "client" in input_variables:
        logger.warning(
            f"'client' was passed both as a parameter for the LLM inference client ('{client}') and in input_variables "
            f"('{input_variables['client']}'). The first parameter version will be used for formatting, "
            "while the second input_variables version will be used in template population."
        )

    prompt = self.populate_template(**input_variables)
    return prompt.format_for_client(client)

to_langchain_template 

to_langchain_template()

Convert the ChatPromptTemplate to a LangChain ChatPromptTemplate.

Examples:

>>> from hf_hub_prompts import download_prompt_template
>>> prompt_template = download_prompt_template(
...     repo_id="MoritzLaurer/example_prompts",
...     filename="code_teacher.yaml"
... )
>>> lc_template = prompt_template.to_langchain_template()
>>> # test equivalence
>>> from langchain_core.prompts import ChatPromptTemplate as LC_ChatPromptTemplate
>>> isinstance(lc_template, LC_ChatPromptTemplate)
True

Returns:

Name Type Description
ChatPromptTemplate ChatPromptTemplate

A LangChain ChatPromptTemplate object.

Raises:

Type Description
ImportError

If LangChain is not installed.
Source code in hf_hub_prompts/prompt_templates.py 
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
def to_langchain_template(self) -> "LC_ChatPromptTemplate":
    """Convert the ChatPromptTemplate to a LangChain ChatPromptTemplate.

    Examples:
        >>> from hf_hub_prompts import download_prompt_template
        >>> prompt_template = download_prompt_template(
        ...     repo_id="MoritzLaurer/example_prompts",
        ...     filename="code_teacher.yaml"
        ... )
        >>> lc_template = prompt_template.to_langchain_template()
        >>> # test equivalence
        >>> from langchain_core.prompts import ChatPromptTemplate as LC_ChatPromptTemplate
        >>> isinstance(lc_template, LC_ChatPromptTemplate)
        True

    Returns:
        ChatPromptTemplate: A LangChain ChatPromptTemplate object.

    Raises:
        ImportError: If LangChain is not installed.
    """
    try:
        from langchain_core.prompts import ChatPromptTemplate as LC_ChatPromptTemplate
    except ImportError as e:
        raise ImportError("LangChain is not installed. Please install it with 'pip install langchain'") from e

    return LC_ChatPromptTemplate(
        messages=[(msg["role"], msg["content"]) for msg in self.messages],
        input_variables=self.input_variables,
        metadata=self.metadata,
    )