crewai-getyoutubetranscript

YouTube transcript tools for CrewAI agents, powered by the GetYouTubeTranscript API. Let your crew read any YouTube video's transcript (optionally with [m:ss] timestamps), search YouTube, and list a channel's videos.

The API fetches transcripts on its own servers, so it works from cloud servers without proxies, and without RequestBlocked / IpBlocked errors.

Install

pip install crewai-getyoutubetranscript

Get an API key at getyoutubetranscript.com/dashboard (free tier included):

export GETYOUTUBETRANSCRIPT_API_KEY=sk_live_...

Or pass api_key="..." to any tool.

Tools

API errors (no captions, invalid video, out of credits) come back to the agent as Error: ... text instead of raising.

Example

from crewai import Agent, Crew, Task
from crewai_getyoutubetranscript import GetYouTubeTranscriptTool, YouTubeSearchTool

researcher = Agent(
    role="Video researcher",
    goal="Find the best YouTube talks on a topic and summarize them with timestamps",
    backstory="You turn long videos into short, cited notes.",
    tools=[YouTubeSearchTool(), GetYouTubeTranscriptTool()],
)

task = Task(
    description="Find a talk about vector databases and summarize its key points with [m:ss] timestamps.",
    expected_output="A bullet list of key points, each with a timestamp.",
    agent=researcher,
)

print(Crew(agents=[researcher], tasks=[task]).kickoff())

Use a tool directly:

print(GetYouTubeTranscriptTool().run(video="https://youtu.be/jNQXAC9IVRw", timestamps=True))

Pricing

Each transcript, search or channel page uses one credit from your GetYouTubeTranscript account. Failed requests are not charged.

Links

• GetYouTubeTranscript API docs
• Python SDK (this package is built on it)
• License: MIT