mirror of https://github.com/crewAIInc/crewAI.git synced 2026-01-14 10:38:29 +00:00

Files

Jakub Strnad 2f80840c74 fix: Ensure tools handle parameters passed post-creation correctly (#3 )

- Fixed an issue where multiple tools failed to function if parameters were provided after tool creation.
- Updated tools to correctly process source file/URL passed by the agent post-creation as per documentation.

Closes #<47>

2024-06-21 15:06:17 +02:00

README.md

Custom model docs

2024-03-21 21:13:36 -03:00

website_search_tool.py

fix: Ensure tools handle parameters passed post-creation correctly (#3 )

2024-06-21 15:06:17 +02:00

README.md

WebsiteSearchTool

Description

This tool is specifically crafted for conducting semantic searches within the content of a particular website. Leveraging a Retrieval-Augmented Generation (RAG) model, it navigates through the information provided on a given URL. Users have the flexibility to either initiate a search across any website known or discovered during its usage or to concentrate the search on a predefined, specific website.

Installation

Install the crewai_tools package by executing the following command in your terminal:

pip install 'crewai[tools]'

Example

To utilize the WebsiteSearchTool for different use cases, follow these examples:

from crewai_tools import WebsiteSearchTool

# To enable the tool to search any website the agent comes across or learns about during its operation
tool = WebsiteSearchTool()

# OR

# To restrict the tool to only search within the content of a specific website.
tool = WebsiteSearchTool(website='https://example.com')

Arguments

website : An optional argument that specifies the valid website URL to perform the search on. This becomes necessary if the tool is initialized without a specific website. In the WebsiteSearchToolSchema, this argument is mandatory. However, in the FixedWebsiteSearchToolSchema, it becomes optional if a website is provided during the tool's initialization, as it will then only search within the predefined website's content.

Custom model and embeddings

By default, the tool uses OpenAI for both embeddings and summarization. To customize the model, you can use a config dictionary as follows:

tool = WebsiteSearchTool(
    config=dict(
        llm=dict(
            provider="ollama", # or google, openai, anthropic, llama2, ...
            config=dict(
                model="llama2",
                # temperature=0.5,
                # top_p=1,
                # stream=true,
            ),
        ),
        embedder=dict(
            provider="google",
            config=dict(
                model="models/embedding-001",
                task_type="retrieval_document",
                # title="Embeddings",
            ),
        ),
    )
)