Skip to content

Index

AsyncBaseTool #

Bases: BaseTool

Base-level tool class that is backwards compatible with the old tool spec but also supports async.

Source code in llama-index-core/llama_index/core/tools/types.py 
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
class AsyncBaseTool(BaseTool):
    """
    Base-level tool class that is backwards compatible with the old tool spec but also
    supports async.
    """

    def __call__(self, *args: Any, **kwargs: Any) -> ToolOutput:
        return self.call(*args, **kwargs)

    @abstractmethod
    def call(self, input: Any) -> ToolOutput:
        """
        This is the method that should be implemented by the tool developer.
        """

    @abstractmethod
    async def acall(self, input: Any) -> ToolOutput:
        """
        This is the async version of the call method.
        Should also be implemented by the tool developer as an
        async-compatible implementation.
        """

call abstractmethod #

call(input: Any) -> ToolOutput

This is the method that should be implemented by the tool developer.

Source code in llama-index-core/llama_index/core/tools/types.py 
160
161
162
163
164
@abstractmethod
def call(self, input: Any) -> ToolOutput:
    """
    This is the method that should be implemented by the tool developer.
    """

acall abstractmethod async #

acall(input: Any) -> ToolOutput

This is the async version of the call method. Should also be implemented by the tool developer as an async-compatible implementation.

Source code in llama-index-core/llama_index/core/tools/types.py 
166
167
168
169
170
171
172
@abstractmethod
async def acall(self, input: Any) -> ToolOutput:
    """
    This is the async version of the call method.
    Should also be implemented by the tool developer as an
    async-compatible implementation.
    """

BaseToolAsyncAdapter #

Bases: AsyncBaseTool

Adapter class that allows a synchronous tool to be used as an async tool.

Source code in llama-index-core/llama_index/core/tools/types.py 
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
class BaseToolAsyncAdapter(AsyncBaseTool):
    """
    Adapter class that allows a synchronous tool to be used as an async tool.
    """

    def __init__(self, tool: BaseTool):
        self.base_tool = tool

    @property
    def metadata(self) -> ToolMetadata:
        return self.base_tool.metadata

    def call(self, input: Any) -> ToolOutput:
        return self.base_tool(input)

    async def acall(self, input: Any) -> ToolOutput:
        return self.call(input)

BaseTool #

Source code in llama-index-core/llama_index/core/tools/types.py 
 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
class BaseTool:
    @property
    @abstractmethod
    def metadata(self) -> ToolMetadata:
        pass

    @abstractmethod
    def __call__(self, input: Any) -> ToolOutput:
        pass

    def _process_langchain_tool_kwargs(
        self,
        langchain_tool_kwargs: Any,
    ) -> Dict[str, Any]:
        """Process langchain tool kwargs."""
        if "name" not in langchain_tool_kwargs:
            langchain_tool_kwargs["name"] = self.metadata.name or ""
        if "description" not in langchain_tool_kwargs:
            langchain_tool_kwargs["description"] = self.metadata.description
        if "fn_schema" not in langchain_tool_kwargs:
            langchain_tool_kwargs["args_schema"] = self.metadata.fn_schema
        return langchain_tool_kwargs

    def to_langchain_tool(
        self,
        **langchain_tool_kwargs: Any,
    ) -> "Tool":
        """To langchain tool."""
        from llama_index.core.bridge.langchain import Tool

        langchain_tool_kwargs = self._process_langchain_tool_kwargs(
            langchain_tool_kwargs
        )
        return Tool.from_function(
            func=self.__call__,
            **langchain_tool_kwargs,
        )

    def to_langchain_structured_tool(
        self,
        **langchain_tool_kwargs: Any,
    ) -> "StructuredTool":
        """To langchain structured tool."""
        from llama_index.core.bridge.langchain import StructuredTool

        langchain_tool_kwargs = self._process_langchain_tool_kwargs(
            langchain_tool_kwargs
        )
        return StructuredTool.from_function(
            func=self.__call__,
            **langchain_tool_kwargs,
        )

to_langchain_tool #

to_langchain_tool(**langchain_tool_kwargs: Any) -> Tool

To langchain tool.

Source code in llama-index-core/llama_index/core/tools/types.py 
120
121
122
123
124
125
126
127
128
129
130
131
132
133
def to_langchain_tool(
    self,
    **langchain_tool_kwargs: Any,
) -> "Tool":
    """To langchain tool."""
    from llama_index.core.bridge.langchain import Tool

    langchain_tool_kwargs = self._process_langchain_tool_kwargs(
        langchain_tool_kwargs
    )
    return Tool.from_function(
        func=self.__call__,
        **langchain_tool_kwargs,
    )

to_langchain_structured_tool #

to_langchain_structured_tool(**langchain_tool_kwargs: Any) -> StructuredTool

To langchain structured tool.

Source code in llama-index-core/llama_index/core/tools/types.py 
135
136
137
138
139
140
141
142
143
144
145
146
147
148
def to_langchain_structured_tool(
    self,
    **langchain_tool_kwargs: Any,
) -> "StructuredTool":
    """To langchain structured tool."""
    from llama_index.core.bridge.langchain import StructuredTool

    langchain_tool_kwargs = self._process_langchain_tool_kwargs(
        langchain_tool_kwargs
    )
    return StructuredTool.from_function(
        func=self.__call__,
        **langchain_tool_kwargs,
    )

ToolMetadata dataclass #

Source code in llama-index-core/llama_index/core/tools/types.py 
18
19
20
21
22
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
@dataclass
class ToolMetadata:
    description: str
    name: Optional[str] = None
    fn_schema: Optional[Type[BaseModel]] = DefaultToolFnSchema
    return_direct: bool = False

    def get_parameters_dict(self) -> dict:
        if self.fn_schema is None:
            parameters = {
                "type": "object",
                "properties": {
                    "input": {"title": "input query string", "type": "string"},
                },
                "required": ["input"],
            }
        else:
            parameters = self.fn_schema.schema()
            parameters = {
                k: v
                for k, v in parameters.items()
                if k in ["type", "properties", "required", "definitions"]
            }
        return parameters

    @property
    def fn_schema_str(self) -> str:
        """Get fn schema as string."""
        if self.fn_schema is None:
            raise ValueError("fn_schema is None.")
        parameters = self.get_parameters_dict()
        return json.dumps(parameters)

    def get_name(self) -> str:
        """Get name."""
        if self.name is None:
            raise ValueError("name is None.")
        return self.name

    @deprecated(
        "Deprecated in favor of `to_openai_tool`, which should be used instead."
    )
    def to_openai_function(self) -> Dict[str, Any]:
        """Deprecated and replaced by `to_openai_tool`.
        The name and arguments of a function that should be called, as generated by the
        model.
        """
        return {
            "name": self.name,
            "description": self.description,
            "parameters": self.get_parameters_dict(),
        }

    def to_openai_tool(self) -> Dict[str, Any]:
        """To OpenAI tool."""
        return {
            "type": "function",
            "function": {
                "name": self.name,
                "description": self.description,
                "parameters": self.get_parameters_dict(),
            },
        }

fn_schema_str property #

fn_schema_str: str

Get fn schema as string.

get_name #

get_name() -> str

Get name.

Source code in llama-index-core/llama_index/core/tools/types.py 
51
52
53
54
55
def get_name(self) -> str:
    """Get name."""
    if self.name is None:
        raise ValueError("name is None.")
    return self.name

to_openai_function #

to_openai_function() -> Dict[str, Any]

Deprecated and replaced by to_openai_tool. The name and arguments of a function that should be called, as generated by the model.

Source code in llama-index-core/llama_index/core/tools/types.py 
57
58
59
60
61
62
63
64
65
66
67
68
69
@deprecated(
    "Deprecated in favor of `to_openai_tool`, which should be used instead."
)
def to_openai_function(self) -> Dict[str, Any]:
    """Deprecated and replaced by `to_openai_tool`.
    The name and arguments of a function that should be called, as generated by the
    model.
    """
    return {
        "name": self.name,
        "description": self.description,
        "parameters": self.get_parameters_dict(),
    }

to_openai_tool #

to_openai_tool() -> Dict[str, Any]

To OpenAI tool.

Source code in llama-index-core/llama_index/core/tools/types.py 
71
72
73
74
75
76
77
78
79
80
def to_openai_tool(self) -> Dict[str, Any]:
    """To OpenAI tool."""
    return {
        "type": "function",
        "function": {
            "name": self.name,
            "description": self.description,
            "parameters": self.get_parameters_dict(),
        },
    }

ToolOutput #

Bases: BaseModel

Tool output.

Source code in llama-index-core/llama_index/core/tools/types.py 
83
84
85
86
87
88
89
90
91
92
93
94
class ToolOutput(BaseModel):
    """Tool output."""

    content: str
    tool_name: str
    raw_input: Dict[str, Any]
    raw_output: Any
    is_error: bool = False

    def __str__(self) -> str:
        """String."""
        return str(self.content)