Implementation:CrewAIInc CrewAI YouTube Video Search Tool

Knowledge Sources	CrewAI
Domains	Tools, RAG, YouTube
Last Updated	2026-02-11 00:00 GMT

Overview

YoutubeVideoSearchTool performs RAG-based semantic search over individual YouTube video transcripts using vector embeddings.

Description

The YoutubeVideoSearchTool extends RagTool and uses the dual-schema pattern: YoutubeVideoSearchToolSchema requires both search_query and youtube_video_url, while FixedYoutubeVideoSearchToolSchema requires only search_query (used when a specific video URL is pre-configured). During initialization, if a youtube_video_url is provided, it is added to the RAG index via the add method (which delegates to RagTool.add with DataType.YOUTUBE_VIDEO), the description is updated, and the schema switches to the fixed variant. The _run method optionally adds a new video URL at runtime before delegating semantic search to the parent RagTool._run with configurable similarity_threshold and limit parameters.

Usage

Use this tool when a CrewAI agent needs to semantically search through YouTube video transcripts, enabling intelligent extraction of specific information from video content for video-based research and analysis workflows.

Code Reference

Source Location

• Repository: CrewAI
• File: lib/crewai-tools/src/crewai_tools/tools/youtube_video_search_tool/youtube_video_search_tool.py
• Lines: 1-51

Signature

class FixedYoutubeVideoSearchToolSchema(BaseModel):
    search_query: str = Field(..., description="Mandatory search query ...")

class YoutubeVideoSearchToolSchema(FixedYoutubeVideoSearchToolSchema):
    youtube_video_url: str = Field(..., description="Mandatory youtube_video_url path you want to search")

class YoutubeVideoSearchTool(RagTool):
    name: str = "Search a Youtube Video content"
    description: str = "A tool that can be used to semantic search a query from a Youtube Video content."
    args_schema: type[BaseModel] = YoutubeVideoSearchToolSchema

    def __init__(self, youtube_video_url: str | None = None, **kwargs):
        ...
    def add(self, youtube_video_url: str) -> None:
        ...
    def _run(self, search_query: str, youtube_video_url: str | None = None,
             similarity_threshold: float | None = None, limit: int | None = None) -> str:
        ...

Import

from crewai_tools import YoutubeVideoSearchTool

I/O Contract

Inputs

Name	Type	Required	Description
search_query	str	Yes	The semantic search query to run against the YouTube video content
youtube_video_url	str	No	YouTube video URL to search (can be set at init or runtime)
similarity_threshold	float	No	Minimum similarity score for results
limit	int	No	Maximum number of results to return

Outputs

Name	Type	Description
_run() returns	str	Semantically relevant text passages from the YouTube video transcript

Usage Examples

Basic Usage

from crewai_tools import YoutubeVideoSearchTool

tool = YoutubeVideoSearchTool(youtube_video_url="https://www.youtube.com/watch?v=dQw4w9WgXcQ")
result = tool._run(search_query="main topic discussed")

Dynamic Video

from crewai_tools import YoutubeVideoSearchTool

tool = YoutubeVideoSearchTool()
result = tool._run(
    search_query="key takeaways",
    youtube_video_url="https://www.youtube.com/watch?v=example123"
)

Related Pages

• Principle:CrewAIInc_CrewAI_Built_In_Tool_Selection