What is AutoGen?
Microsoft's framework for building multi-agent conversational AI systems. Agents can converse, use tools, and collaborate autonomously.
Searching YouTube with AutoGen
This integration lets your AutoGen agent search YouTube in real time via the Scavio API. The agent gets back structured JSON with video results, channel results, playlist results, transcripts -- ready for reasoning and decision-making.
Setup
Bash
pip install autogen-agentchat requestsCode Example
Here is a complete AutoGen agent that searches YouTube using Scavio:
Python
from autogen import ConversableAgent
import json
import os
import requests
# YouTube takes search (not query) in the request body.
def scavio_search(query: str) -> str:
"""Query YouTube through the Scavio API (POST /api/v1/youtube/search)."""
response = requests.post(
"https://api.scavio.dev/api/v1/youtube/search",
headers={
"Authorization": "Bearer " + os.environ["SCAVIO_API_KEY"],
"Content-Type": "application/json",
},
json={"search": query},
timeout=60,
)
response.raise_for_status()
return json.dumps(response.json())
assistant = ConversableAgent(
"assistant",
llm_config={"model": "gpt-5.5"},
system_message="You are a helpful research assistant.",
)
user = ConversableAgent("user", human_input_mode="NEVER")
assistant.register_for_llm(name="search", description="Query YouTube")(scavio_search)
user.register_for_execution(name="search")(scavio_search)
user.initiate_chat(assistant, message="Research: python web scraping tutorial")Full Working Example
A production-ready example with error handling:
Python
from autogen import ConversableAgent
import json
import os
import requests
# YouTube takes search (not query) in the request body.
def scavio_search(query: str) -> str:
"""Query YouTube through the Scavio API (POST /api/v1/youtube/search)."""
response = requests.post(
"https://api.scavio.dev/api/v1/youtube/search",
headers={
"Authorization": "Bearer " + os.environ["SCAVIO_API_KEY"],
"Content-Type": "application/json",
},
json={"search": query},
timeout=60,
)
response.raise_for_status()
return json.dumps(response.json())
assistant = ConversableAgent(
"assistant",
llm_config={"model": "gpt-5.5"},
system_message="You are a helpful research assistant with real-time search.",
)
user = ConversableAgent("user", human_input_mode="NEVER", max_consecutive_auto_reply=3)
assistant.register_for_llm(name="search", description="Query YouTube")(scavio_search)
user.register_for_execution(name="search")(scavio_search)
user.initiate_chat(assistant, message="Research: python web scraping tutorial")Pricing
Scavio offers a free tier with 50 credits on signup (1 credit per search). No credit card required. This is enough to build and test your AutoGen integration. Paid plans start at $30/month for higher volumes.