toolregistry 0.4.3

A library for managing tool registries

ToolRegistry

中文版

A Python library for managing and executing tools in a structured way.

Features

• Tool registration and management
• JSON Schema generation for tool parameters
• Tool execution and result handling
• Support for both synchronous and asynchronous tools
• Support MCP sse, OpenAPI tools

Full Documentation

Full documentation is available at https://toolregistry.lab.oaklight.cn

Installation

Basic Installation

Install the core package (requires Python >= 3.8):

pip install toolregistry

Installing with Extra Support Modules

Extra modules can be installed by specifying extras in brackets. For example, to install specific extra supports:

pip install toolregistry[mcp,openapi]

Below is a table summarizing available extra modules:

Extra Module	Python Requirement	Example Command
mcp	Python >= 3.10	pip install toolregistry[mcp]
openapi	Python >= 3.8	pip install toolregistry[openapi]

Examples

OpenAI Implementation

The openai_tool_usage_example.py shows how to integrate ToolRegistry with OpenAI's API.

Cicada Implementation

The cicada_tool_usage_example.py demonstrates how to use ToolRegistry with the Cicada MultiModalModel.

Basic Tool Invocation

This section demonstrates how to invoke a basic tool. Example:

from toolregistry import ToolRegistry

registry = ToolRegistry()

@registry.register
def add(a: float, b: float) -> float:
    """Add two numbers together."""
    return a + b

available_tools = registry.get_available_tools()

print(available_tools) # ['add']

add_func = registry.get_callable('add')
print(type(add_func)) # <class 'function'>
add_result = add_func(1, 2)
print(add_result) # 3

add_func = registry['add']
print(type(add_func)) # <class 'function'>
add_result = add_func(4, 5)
print(add_result) # 9

For more usage examples, please refer to Documentation - Usage

MCP Integration

The ToolRegistry provides first-class support for MCP (Model Context Protocol) tools:

registry.register_mcp_tools("http://localhost:8000/mcp/sse")

# Get all tools JSON including MCP tools
tools_json = registry.get_tools_json()

OpenAPI Integration

ToolRegistry supports integration with OpenAPI for interacting with tools using a standardized API interface:

registry.register_openapi_tools("http://localhost:8000/") # by providing baseurl
registry.register_openapi_tools("./openapi_spec.json", "http://localhost/") # by providing local OpenAPI spec file and base url

# Get all tools JSON including OpenAPI tools
tools_json = registry.get_tools_json()

Static Method Integration and Hub of Tools

ToolRegistry supports integration of static methods and predefined hub tools for enhanced functionality and organization.

Static Method Integration

Static methods from Python classes can be registered as tools using the StaticMethodIntegration module. This allows developers to extend ToolRegistry by creating custom tool classes with reusable static methods.

Example:

from toolregistry import ToolRegistry

class CustomTools:
    @staticmethod
    def greet(name: str) -> str:
        return f"Hello, {name}!"

registry = ToolRegistry()
registry.register_static_tools(CustomTools) 

# List registered tools
print(registry.get_available_tools())
# Output: ['greet']

Hub Tools

Latest Available Tools

Hub tools encapsulate commonly used functionalities as static methods in classes. These tools are grouped for better organization and reusability.

Examples of available hub tools include:

• Calculator: Basic arithmetic, scientific operations, statistical functions, financial calculations, and more.
• FileOps: File manipulation operations like diff generation, patching, and verification.
• Filesystem: Comprehensive file system operations such as directory listing, file reading/writing, and path manipulation.
• UnitConverter: Extensive unit conversion tools for temperature, length, weight, and more.

To register hub tools:

from toolregistry import ToolRegistry
from toolregistry.hub import Calculator

registry = ToolRegistry()
registry.register_static_tools(Calculator, with_namespace=True)

# Get available tools list
print(registry.get_available_tools())
# Output: ['Calculator.add', 'Calculator.subtract', ..., 'Calculator.multiply', ...]

License

This project is licensed under the MIT License - see the LICENSE file for details.