Unified abstractions to represent tools across the framework.
Module tool
Tool
Data class representing a Tool that Language Models can prepare a call for.
Accurate definitions of the textual attributes such as name
and description
are important for the Language Model to correctly prepare the call.
Arguments:
name
: Name of the Tool.description
: Description of the Tool.parameters
: A JSON schema defining the parameters expected by the Tool.function
: The function that will be invoked when the Tool is called.outputs_to_string
: Optional dictionary defining how a tool outputs should be converted into a string. If the source is provided only the specified output key is sent to the handler. If the source is omitted the whole tool result is sent to the handler. Example: { "source": "docs", "handler": format_documents }inputs_from_state
: Optional dictionary mapping state keys to tool parameter names. Example: {"repository": "repo"} maps state's "repository" to tool's "repo" parameter.outputs_to_state
: Optional dictionary defining how tool outputs map to keys within state as well as optional handlers. If the source is provided only the specified output key is sent to the handler. Example: { "documents": {"source": "docs", "handler": custom_handler} } If the source is omitted the whole tool result is sent to the handler. Example: { "documents": {"handler": custom_handler} }
Tool.tool_spec
@property
def tool_spec() -> Dict[str, Any]
Return the Tool specification to be used by the Language Model.
Tool.invoke
def invoke(**kwargs) -> Any
Invoke the Tool with the provided keyword arguments.
Tool.to_dict
def to_dict() -> Dict[str, Any]
Serializes the Tool to a dictionary.
Returns:
Dictionary with serialized data.
Tool.from_dict
@classmethod
def from_dict(cls, data: Dict[str, Any]) -> "Tool"
Deserializes the Tool from a dictionary.
Arguments:
data
: Dictionary to deserialize from.
Returns:
Deserialized Tool.
deserialize_tools_inplace
def deserialize_tools_inplace(data: Dict[str, Any], key: str = "tools")
Deserialize a list of Tools or a Toolset in a dictionary inplace.
Deprecated in favor of deserialize_tools_or_toolset_inplace
. It will be removed in Haystack 2.14.0.
Arguments:
data
: The dictionary with the serialized data.key
: The key in the dictionary where the list of Tools or Toolset is stored.
Module from_function
create_tool_from_function
def create_tool_from_function(
function: Callable,
name: Optional[str] = None,
description: Optional[str] = None,
inputs_from_state: Optional[Dict[str, str]] = None,
outputs_to_state: Optional[Dict[str, Dict[str,
Any]]] = None) -> "Tool"
Create a Tool instance from a function.
Allows customizing the Tool name and description.
For simpler use cases, consider using the @tool
decorator.
Usage example
from typing import Annotated, Literal
from haystack.tools import create_tool_from_function
def get_weather(
city: Annotated[str, "the city for which to get the weather"] = "Munich",
unit: Annotated[Literal["Celsius", "Fahrenheit"], "the unit for the temperature"] = "Celsius"):
'''A simple function to get the current weather for a location.'''
return f"Weather report for {city}: 20 {unit}, sunny"
tool = create_tool_from_function(get_weather)
print(tool)
>>> Tool(name='get_weather', description='A simple function to get the current weather for a location.',
>>> parameters={
>>> 'type': 'object',
>>> 'properties': {
>>> 'city': {'type': 'string', 'description': 'the city for which to get the weather', 'default': 'Munich'},
>>> 'unit': {
>>> 'type': 'string',
>>> 'enum': ['Celsius', 'Fahrenheit'],
>>> 'description': 'the unit for the temperature',
>>> 'default': 'Celsius',
>>> },
>>> }
>>> },
>>> function=<function get_weather at 0x7f7b3a8a9b80>)
Arguments:
function
: The function to be converted into a Tool. The function must include type hints for all parameters. The function is expected to have basic python input types (str, int, float, bool, list, dict, tuple). Other input types may work but are not guaranteed. If a parameter is annotated usingtyping.Annotated
, its metadata will be used as parameter description.name
: The name of the Tool. If not provided, the name of the function will be used.description
: The description of the Tool. If not provided, the docstring of the function will be used. To intentionally leave the description empty, pass an empty string.inputs_from_state
: Optional dictionary mapping state keys to tool parameter names. Example: {"repository": "repo"} maps state's "repository" to tool's "repo" parameter.outputs_to_state
: Optional dictionary defining how tool outputs map to state and message handling. Example: { "documents": {"source": "docs", "handler": custom_handler}, "message": {"source": "summary", "handler": format_summary} }
Raises:
ValueError
: If any parameter of the function lacks a type hint.SchemaGenerationError
: If there is an error generating the JSON schema for the Tool.
Returns:
The Tool created from the function.
tool
def tool(
function: Optional[Callable] = None,
*,
name: Optional[str] = None,
description: Optional[str] = None,
inputs_from_state: Optional[Dict[str, str]] = None,
outputs_to_state: Optional[Dict[str, Dict[str, Any]]] = None
) -> Union[Tool, Callable[[Callable], Tool]]
Decorator to convert a function into a Tool.
Can be used with or without parameters: @tool # without parameters def my_function(): ...
@tool(name="custom_name") # with parameters def my_function(): ...
Usage example
from typing import Annotated, Literal
from haystack.tools import tool
@tool
def get_weather(
city: Annotated[str, "the city for which to get the weather"] = "Munich",
unit: Annotated[Literal["Celsius", "Fahrenheit"], "the unit for the temperature"] = "Celsius"):
'''A simple function to get the current weather for a location.'''
return f"Weather report for {city}: 20 {unit}, sunny"
print(get_weather)
>>> Tool(name='get_weather', description='A simple function to get the current weather for a location.',
>>> parameters={
>>> 'type': 'object',
>>> 'properties': {
>>> 'city': {'type': 'string', 'description': 'the city for which to get the weather', 'default': 'Munich'},
>>> 'unit': {
>>> 'type': 'string',
>>> 'enum': ['Celsius', 'Fahrenheit'],
>>> 'description': 'the unit for the temperature',
>>> 'default': 'Celsius',
>>> },
>>> }
>>> },
>>> function=<function get_weather at 0x7f7b3a8a9b80>)
Arguments:
function
: The function to decorate (when used without parameters)name
: Optional custom name for the tooldescription
: Optional custom descriptioninputs_from_state
: Optional dictionary mapping state keys to tool parameter namesoutputs_to_state
: Optional dictionary defining how tool outputs map to state and message handling
Returns:
Either a Tool instance or a decorator function that will create one
Module component_tool
ComponentTool
A Tool that wraps Haystack components, allowing them to be used as tools by LLMs.
ComponentTool automatically generates LLM-compatible tool schemas from component input sockets,
which are derived from the component's run
method signature and type hints.
Key features:
- Automatic LLM tool calling schema generation from component input sockets
- Type conversion and validation for component inputs
- Support for types:
- Dataclasses
- Lists of dataclasses
- Basic types (str, int, float, bool, dict)
- Lists of basic types
- Automatic name generation from component class name
- Description extraction from component docstrings
To use ComponentTool, you first need a Haystack component - either an existing one or a new one you create. You can create a ComponentTool from the component by passing the component to the ComponentTool constructor. Below is an example of creating a ComponentTool from an existing SerperDevWebSearch component.
from haystack import component, Pipeline
from haystack.tools import ComponentTool
from haystack.components.websearch import SerperDevWebSearch
from haystack.utils import Secret
from haystack.components.tools.tool_invoker import ToolInvoker
from haystack.components.generators.chat import OpenAIChatGenerator
from haystack.dataclasses import ChatMessage
# Create a SerperDev search component
search = SerperDevWebSearch(api_key=Secret.from_env_var("SERPERDEV_API_KEY"), top_k=3)
# Create a tool from the component
tool = ComponentTool(
component=search,
name="web_search", # Optional: defaults to "serper_dev_web_search"
description="Search the web for current information on any topic" # Optional: defaults to component docstring
)
# Create pipeline with OpenAIChatGenerator and ToolInvoker
pipeline = Pipeline()
pipeline.add_component("llm", OpenAIChatGenerator(model="gpt-4o-mini", tools=[tool]))
pipeline.add_component("tool_invoker", ToolInvoker(tools=[tool]))
# Connect components
pipeline.connect("llm.replies", "tool_invoker.messages")
message = ChatMessage.from_user("Use the web search tool to find information about Nikola Tesla")
# Run pipeline
result = pipeline.run({"llm": {"messages": [message]}})
print(result)
ComponentTool.__init__
def __init__(component: Component,
name: Optional[str] = None,
description: Optional[str] = None,
parameters: Optional[Dict[str, Any]] = None,
*,
outputs_to_string: Optional[Dict[str,
Union[str,
Callable[[Any],
str]]]] = None,
inputs_from_state: Optional[Dict[str, str]] = None,
outputs_to_state: Optional[Dict[str,
Dict[str,
Union[str,
Callable]]]] = None)
Create a Tool instance from a Haystack component.
Arguments:
component
: The Haystack component to wrap as a tool.name
: Optional name for the tool (defaults to snake_case of component class name).description
: Optional description (defaults to component's docstring).parameters
: A JSON schema defining the parameters expected by the Tool. Will fall back to the parameters defined in the component's run method signature if not provided.outputs_to_string
: Optional dictionary defining how a tool outputs should be converted into a string. If the source is provided only the specified output key is sent to the handler. If the source is omitted the whole tool result is sent to the handler. Example: { "source": "docs", "handler": format_documents }inputs_from_state
: Optional dictionary mapping state keys to tool parameter names. Example: {"repository": "repo"} maps state's "repository" to tool's "repo" parameter.outputs_to_state
: Optional dictionary defining how tool outputs map to keys within state as well as optional handlers. If the source is provided only the specified output key is sent to the handler. Example: { "documents": {"source": "docs", "handler": custom_handler} } If the source is omitted the whole tool result is sent to the handler. Example: { "documents": {"handler": custom_handler} }
Raises:
ValueError
: If the component is invalid or schema generation fails.
ComponentTool.to_dict
def to_dict() -> Dict[str, Any]
Serializes the ComponentTool to a dictionary.
ComponentTool.from_dict
@classmethod
def from_dict(cls, data: Dict[str, Any]) -> "Tool"
Deserializes the ComponentTool from a dictionary.
ComponentTool.tool_spec
@property
def tool_spec() -> Dict[str, Any]
Return the Tool specification to be used by the Language Model.
ComponentTool.invoke
def invoke(**kwargs) -> Any
Invoke the Tool with the provided keyword arguments.
Module toolset
Toolset
A collection of related Tools that can be used and managed as a cohesive unit.
Toolset serves two main purposes:
- Group related tools together: Toolset allows you to organize related tools into a single collection, making it easier to manage and use them as a unit in Haystack pipelines.
Example:
from haystack.tools import Tool, Toolset
from haystack.components.tools import ToolInvoker
# Define math functions
def add_numbers(a: int, b: int) -> int:
return a + b
def subtract_numbers(a: int, b: int) -> int:
return a - b
# Create tools with proper schemas
add_tool = Tool(
name="add",
description="Add two numbers",
parameters={
"type": "object",
"properties": {
"a": {"type": "integer"},
"b": {"type": "integer"}
},
"required": ["a", "b"]
},
function=add_numbers
)
subtract_tool = Tool(
name="subtract",
description="Subtract b from a",
parameters={
"type": "object",
"properties": {
"a": {"type": "integer"},
"b": {"type": "integer"}
},
"required": ["a", "b"]
},
function=subtract_numbers
)
# Create a toolset with the math tools
math_toolset = Toolset([add_tool, subtract_tool])
# Use the toolset with a ToolInvoker or ChatGenerator component
invoker = ToolInvoker(tools=math_toolset)
- Base class for dynamic tool loading: By subclassing Toolset, you can create implementations that dynamically load tools from external sources like OpenAPI URLs, MCP servers, or other resources.
Example:
from haystack.core.serialization import generate_qualified_class_name
from haystack.tools import Tool, Toolset
from haystack.components.tools import ToolInvoker
class CalculatorToolset(Toolset):
'''A toolset for calculator operations.'''
def __init__(self):
tools = self._create_tools()
super().__init__(tools)
def _create_tools(self):
# These Tool instances are obviously defined statically and for illustration purposes only.
# In a real-world scenario, you would dynamically load tools from an external source here.
tools = []
add_tool = Tool(
name="add",
description="Add two numbers",
parameters={
"type": "object",
"properties": {"a": {"type": "integer"}, "b": {"type": "integer"}},
"required": ["a", "b"],
},
function=lambda a, b: a + b,
)
multiply_tool = Tool(
name="multiply",
description="Multiply two numbers",
parameters={
"type": "object",
"properties": {"a": {"type": "integer"}, "b": {"type": "integer"}},
"required": ["a", "b"],
},
function=lambda a, b: a * b,
)
tools.append(add_tool)
tools.append(multiply_tool)
return tools
def to_dict(self):
return {
"type": generate_qualified_class_name(type(self)),
"data": {}, # no data to serialize as we define the tools dynamically
}
@classmethod
def from_dict(cls, data):
return cls() # Recreate the tools dynamically during deserialization
# Create the dynamic toolset and use it with ToolInvoker
calculator_toolset = CalculatorToolset()
invoker = ToolInvoker(tools=calculator_toolset)
Toolset implements the collection interface (iter, contains, len, getitem), making it behave like a list of Tools. This makes it compatible with components that expect iterable tools, such as ToolInvoker or Haystack chat generators.
When implementing a custom Toolset subclass for dynamic tool loading:
- Perform the dynamic loading in the init method
- Override to_dict() and from_dict() methods if your tools are defined dynamically
- Serialize endpoint descriptors rather than tool instances if your tools are loaded from external sources
Toolset.__post_init__
def __post_init__()
Validate and set up the toolset after initialization.
This handles the case when tools are provided during initialization.
Toolset.__iter__
def __iter__() -> Iterator[Tool]
Return an iterator over the Tools in this Toolset.
This allows the Toolset to be used wherever a list of Tools is expected.
Returns:
An iterator yielding Tool instances
Toolset.__contains__
def __contains__(item: Any) -> bool
Check if a tool is in this Toolset.
Supports checking by:
- Tool instance: tool in toolset
- Tool name: "tool_name" in toolset
Arguments:
item
: Tool instance or tool name string
Returns:
True if contained, False otherwise
Toolset.add
def add(tool: Union[Tool, "Toolset"]) -> None
Add a new Tool or merge another Toolset.
Arguments:
tool
: A Tool instance or another Toolset to add
Raises:
ValueError
: If adding the tool would result in duplicate tool namesTypeError
: If the provided object is not a Tool or Toolset
Toolset.to_dict
def to_dict() -> Dict[str, Any]
Serialize the Toolset to a dictionary.
Returns:
A dictionary representation of the Toolset Note for subclass implementers: The default implementation is ideal for scenarios where Tool resolution is static. However, if your subclass of Toolset dynamically resolves Tool instances from external sources—such as an MCP server, OpenAPI URL, or a local OpenAPI specification—you should consider serializing the endpoint descriptor instead of the Tool instances themselves. This strategy preserves the dynamic nature of your Toolset and minimizes the overhead associated with serializing potentially large collections of Tool objects. Moreover, by serializing the descriptor, you ensure that the deserialization process can accurately reconstruct the Tool instances, even if they have been modified or removed since the last serialization. Failing to serialize the descriptor may lead to issues where outdated or incorrect Tool configurations are loaded, potentially causing errors or unexpected behavior.
Toolset.from_dict
@classmethod
def from_dict(cls, data: Dict[str, Any]) -> "Toolset"
Deserialize a Toolset from a dictionary.
Arguments:
data
: Dictionary representation of the Toolset
Returns:
A new Toolset instance
Toolset.__add__
def __add__(other: Union[Tool, "Toolset", List[Tool]]) -> "Toolset"
Concatenate this Toolset with another Tool, Toolset, or list of Tools.
Arguments:
other
: Another Tool, Toolset, or list of Tools to concatenate
Raises:
TypeError
: If the other parameter is not a Tool, Toolset, or list of ToolsValueError
: If the combination would result in duplicate tool names
Returns:
A new Toolset containing all tools
Toolset.__len__
def __len__() -> int
Return the number of Tools in this Toolset.
Returns:
Number of Tools
Toolset.__getitem__
def __getitem__(index)
Get a Tool by index.
Arguments:
index
: Index of the Tool to get
Returns:
The Tool at the specified index