Tools

Tip

Location within the framework bee-agent-framework/tools.

Tools in the context of an agent refer to additional functionalities or capabilities integrated with the agent to perform specific tasks beyond text processing.

These tools extend the agent's abilities, allowing it to interact with external systems, access information, and execute actions.

Built-in tools

Name	Description
`PythonTool`	Run arbitrary Python code in the remote environment.
`WikipediaTool`	Search for data on Wikipedia.
`GoogleSearchTool`	Search for data on Google using Custom Search Engine.
`DuckDuckGoTool`	Search for data on DuckDuckGo.
`SQLTool`	Execute SQL queries against relational databases.
`ElasticSearchTool`	Perform search or aggregation queries against an ElasticSearch database.
`CustomTool`	Run your own Python function in the remote environment.
`LLMTool`	Use an LLM to process input data.
`DynamicTool`	Construct to create dynamic tools.
`ArXivTool`	Retrieve research articles published on arXiv.
`WebCrawlerTool`	Retrieve content of an arbitrary website.
`OpenMeteoTool`	Retrieve current, previous, or upcoming weather for a given destination.
`MilvusDatabaseTool`	Perform retrieval queries (search, insert, delete, manage collections) against a MilvusDatabaseTool database.
➕ Request

All examples can be found here.

Tip

Would you like to use a tool from LangChain? See the example.

Usage

Basic

import { OpenMeteoTool } from "bee-agent-framework/tools/weather/openMeteo";

const tool = new OpenMeteoTool();
const result = await tool.run({
  location: { name: "New York" },
  start_date: "2024-10-10",
  end_date: "2024-10-10",
});
console.log(result.getTextContent());

Source: examples/tools/base.ts

Advanced

import { OpenMeteoTool } from "bee-agent-framework/tools/weather/openMeteo";
import { UnconstrainedCache } from "bee-agent-framework/cache/unconstrainedCache";

const tool = new OpenMeteoTool({
  cache: new UnconstrainedCache(),
  retryOptions: {
    maxRetries: 3,
  },
});
console.log(tool.name); // OpenMeteo
console.log(tool.description); // Retrieve current, past, or future weather forecasts for a location.
console.log(tool.inputSchema()); // (zod/json schema)

await tool.cache.clear();

const result = await tool.run({
  location: { name: "New York" },
  start_date: "2024-10-10",
  end_date: "2024-10-10",
  temperature_unit: "celsius",
});
console.log(result.isEmpty()); // false
console.log(result.result); // prints raw data
console.log(result.getTextContent()); // prints data as text

Source: examples/tools/advanced.ts

Tip

To learn more about caching, refer to the Cache documentation page.

Usage with agents

import { OllamaChatLLM } from "bee-agent-framework/adapters/ollama/chat";
import { ArXivTool } from "bee-agent-framework/tools/arxiv";
import { BeeAgent } from "bee-agent-framework/agents/bee/agent";
import { UnconstrainedMemory } from "bee-agent-framework/memory/unconstrainedMemory";

const agent = new BeeAgent({
  llm: new OllamaChatLLM(),
  memory: new UnconstrainedMemory(),
  tools: [new ArXivTool()],
});

Source: examples/tools/agent.ts

Writing a new tool

To create a new tool, you have the following options on how to do that:

Implement the base Tool class.
Initiate the DynamicTool by passing your own handler (function) with the name, description and input schema.
Initiate the CustomTool by passing your own Python function (code interpreter needed).

Implementing the `Tool` class

The recommended and most sustainable way to create a tool is by implementing the base Tool class.

Basic

import {
  ToolEmitter,
  StringToolOutput,
  Tool,
  ToolInput,
  ToolInputValidationError,
} from "bee-agent-framework/tools/base";
import { z } from "zod";
import { randomInteger } from "remeda";
import { Emitter } from "bee-agent-framework/emitter/emitter";

export class RiddleTool extends Tool<StringToolOutput> {
  name = "Riddle";
  description = "It generates a random puzzle to test your knowledge.";

  public readonly emitter: ToolEmitter<ToolInput<this>, StringToolOutput> = Emitter.root.child({
    namespace: ["tool", "riddle"],
    creator: this,
  });

  inputSchema() {
    return z.object({
      index: z
        .number()
        .int()
        .min(0)
        .max(RiddleTool.data.length - 1)
        .optional(),
    });
  }

  public static data = [
    "What has hands but can’t clap?",
    "What has a face and two hands but no arms or legs?",
    "What gets wetter the more it dries?",
    "What has to be broken before you can use it?",
    "What has a head, a tail, but no body?",
    "The more you take, the more you leave behind. What am I?",
    "What goes up but never comes down?",
  ];

  static {
    // Makes the class serializable
    this.register();
  }

  protected async _run(input: ToolInput<this>): Promise<StringToolOutput> {
    const index = input.index ?? randomInteger(0, RiddleTool.data.length - 1);
    const riddle = RiddleTool.data[index];
    if (!riddle) {
      throw new ToolInputValidationError(`Riddle with such index (${index}) does not exist!`);
    }
    return new StringToolOutput(riddle);
  }
}

Source: examples/tools/custom/base.ts

Tip

inputSchema can be asynchronous.

Tip

If you want to return an array or a plain object, use JSONToolOutput or implement your own.

Advanced

If your tool is more complex, you may want to use the full power of the tool abstraction, as the following example shows.

import {
  BaseToolOptions,
  BaseToolRunOptions,
  Tool,
  ToolInput,
  JSONToolOutput,
  ToolError,
  ToolEmitter,
} from "bee-agent-framework/tools/base";
import { z } from "zod";
import { createURLParams } from "bee-agent-framework/internals/fetcher";
import { GetRunContext } from "bee-agent-framework/context";
import { Callback, Emitter } from "bee-agent-framework/emitter/emitter";

type ToolOptions = BaseToolOptions & { maxResults?: number };
type ToolRunOptions = BaseToolRunOptions;

export interface OpenLibraryResponse {
  numFound: number;
  start: number;
  numFoundExact: boolean;
  q: string;
  offset: number;
  docs: Record<string, any>[];
}

export class OpenLibraryToolOutput extends JSONToolOutput<OpenLibraryResponse> {
  isEmpty(): boolean {
    return !this.result || this.result.numFound === 0 || this.result.docs.length === 0;
  }
}

export class OpenLibraryTool extends Tool<OpenLibraryToolOutput, ToolOptions, ToolRunOptions> {
  name = "OpenLibrary";
  description =
    "Provides access to a library of books with information about book titles, authors, contributors, publication dates, publisher and isbn.";

  inputSchema() {
    return z
      .object({
        title: z.string(),
        author: z.string(),
        isbn: z.string(),
        subject: z.string(),
        place: z.string(),
        person: z.string(),
        publisher: z.string(),
      })
      .partial();
  }

  public readonly emitter: ToolEmitter<
    ToolInput<this>,
    OpenLibraryToolOutput,
    {
      beforeFetch: Callback<{ request: { url: string; options: RequestInit } }>;
      afterFetch: Callback<{ data: OpenLibraryResponse }>;
    }
  > = Emitter.root.child({
    namespace: ["tool", "search", "openLibrary"],
    creator: this,
  });

  static {
    this.register();
  }

  protected async _run(
    input: ToolInput<this>,
    _options: Partial<ToolRunOptions>,
    run: GetRunContext<this>,
  ) {
    const request = {
      url: `https://openlibrary.org?${createURLParams({
        searchon: input,
      })}`,
      options: { signal: run.signal } as RequestInit,
    };

    await run.emitter.emit("beforeFetch", { request });
    const response = await fetch(request.url, request.options);

    if (!response.ok) {
      throw new ToolError(
        "Request to Open Library API has failed!",
        [new Error(await response.text())],
        {
          context: { input },
        },
      );
    }

    const json: OpenLibraryResponse = await response.json();
    if (this.options.maxResults) {
      json.docs.length = this.options.maxResults;
    }

    await run.emitter.emit("afterFetch", { data: json });
    return new OpenLibraryToolOutput(json);
  }
}

Source: examples/tools/custom/openLibrary.ts

Implementation Notes

Implement the Tool class:
- MyNewToolOutput is required, must be an implementation of ToolOutput such as StringToolOutput or JSONToolOutput.
- ToolOptions is optional (default BaseToolOptions), constructor parameters that are passed during tool creation
- ToolRunOptions is optional (default BaseToolRunOptions), optional parameters that are passed to the run method
Be given a unique name:

Note: Convention and best practice is to set the tool's name to the name of its class
```
name = "MyNewTool";
```
Provide a natural language description of what the tool does:

❗Important: the agent uses this description to determine when the tool should be used. It's probably the most important aspect of your tool and you should experiment with different natural language descriptions to ensure the tool is used in the correct circumstances. You can also include usage tips and guidance for the agent in the description, but its advisable to keep the description succinct in order to reduce the probability of conflicting with other tools, or adversely affecting agent behavior.
```
description = "Takes X action when given Y input resulting in Z output";
```
Declare an input schema:

This is used to define the format of the input to your tool. The agent will formalise the natural language input(s) it has received and structure them into the fields described in the tool's input. The input schema can be specified using Zod (recommended) or JSONSchema. It must be a function (either sync or async). Zod effects (e.g. z.object().transform(...)) are not supported. The return value of inputSchema must always be an object and pass validation by the validateSchema() function defined in schema.ts. Keep your tool input schema simple and provide schema descriptions to help the agent to interpret fields.
```
inputSchema() {
    // any Zod definition is good here, this is typical simple example
    return z.object({
      // list of key-value pairs
      expression: z
      .string()
      .min(1)
      .describe(
        `The mathematical expression to evaluate (e.g., "2 + 3 * 4").`,
      ),
    });
}
```
Implement initialisation:

The unnamed static block is executed when your tool is called for the first time. It is used to register your tool as serializable (you can then use the serialize() method).
```
static {
    this.register();
}
```

Implement the _run() method:

protected async _run(input: ToolInput<this>, options: Partial<BaseToolRunOptions>, run: RunContext<this>) {
    // insert custom code here
    // MUST: return an instance of the output type specified in the tool class definition
    // MAY: throw an instance of ToolError upon unrecoverable error conditions encountered by the tool
}

Using the `DynamicTool` class

The DynamicTool allows you to create a tool without extending the base tool class.

import { DynamicTool, StringToolOutput } from "bee-agent-framework/tools/base";
import { z } from "zod";

const tool = new DynamicTool({
  name: "GenerateRandomNumber",
  description: "Generates a random number in the given interval.",
  inputSchema: z.object({
    min: z.number().int().min(0),
    max: z.number().int(),
  }),
  async handler(input) {
    const min = Math.min(input.min, input.max);
    const max = Math.max(input.max, input.min);

    const number = Math.floor(Math.random() * (max - min + 1)) + min;
    return new StringToolOutput(number.toString());
  },
});

Source: examples/tools/custom/dynamic.ts

The name of the tool is required and must only contain characters between a-z, A-Z, 0-9, or one of - or _. The inputSchema and description are also both required.

Using the `CustomTool` (Python functions)

If you want to use the Python function, use the CustomTool.

import "dotenv/config";
import { CustomTool } from "bee-agent-framework/tools/custom";

const customTool = await CustomTool.fromSourceCode(
  {
    // Ensure the env exists
    url: process.env.CODE_INTERPRETER_URL!,
  },
  `import requests
from typing import Optional, Union, Dict

def get_riddle() -> Optional[Dict[str, str]]:
  """
  Fetches a random riddle from the Riddles API.

  This function retrieves a random riddle and its answer. It does not accept any input parameters.

  Returns:
      Optional[Dict[str, str]]: A dictionary containing:
          - 'riddle' (str): The riddle question.
          - 'answer' (str): The answer to the riddle.
      Returns None if the request fails.
  """
  url = 'https://riddles-api.vercel.app/random'
  
  try:
      response = requests.get(url)
      response.raise_for_status() 
      return response.json() 
  except Exception as e:
      return None`,
);

Source: examples/tools/custom/python.ts

Important

Custom tools are executed within the code interpreter, but they cannot access any files. Only PythonTool does.

General Tips

Data Minimization

If your tool is providing data to the agent, try to ensure that the data is relevant and free of extraneous metatdata. Preprocessing data to improve relevance and minimize unnecessary data conserves agent memory, improving overall performance.

Provide Hints

If your tool encounters an error that is fixable, you can return a hint to the agent; the agent will try to reuse the tool in the context of the hint. This can improve the agent's ability to recover from errors.

Security & Stability

When building tools, consider that the tool is being invoked by a somewhat unpredictable third party (the agent). You should ensure that sufficient guardrails are in place to prevent adverse outcomes.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

tools.md

tools.md

Tools

Built-in tools

Usage

Basic

Advanced

Usage with agents

Writing a new tool

Implementing the `Tool` class

Basic

Advanced

Implementation Notes

Using the `DynamicTool` class

Using the `CustomTool` (Python functions)

General Tips

Data Minimization

Provide Hints

Security & Stability

Files

tools.md

Latest commit

History

tools.md

File metadata and controls

Tools

Built-in tools

Usage

Basic

Advanced

Usage with agents

Writing a new tool

Implementing the Tool class

Basic

Advanced

Implementation Notes

Using the DynamicTool class

Using the CustomTool (Python functions)

General Tips

Data Minimization

Provide Hints

Security & Stability

Implementing the `Tool` class

Using the `DynamicTool` class

Using the `CustomTool` (Python functions)