File size: 1,691 Bytes
27ecee0 9b5b26a c19d193 8fe992b 27ecee0 9b5b26a 27ecee0 9b5b26a 27ecee0 9b5b26a 27ecee0 8c01ffb 27ecee0 e121372 27ecee0 13d500a 8c01ffb 27ecee0 861422e 27ecee0 8c01ffb 8fe992b 27ecee0 8c01ffb 861422e 8fe992b 27ecee0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 |
from smolagents import CodeAgent, HfApiModel, tool
import requests
import yaml
# Define a research tool using EXA API
@tool
def search_cape_town_weather() -> str:
"""Searches for the latest weather in Cape Town using EXA search API."""
api_key = "fe5ee653-7f0d-493c-99ae-8e63d46133d9" # Replace with your free EXA API key
query = "current weather in Cape Town site:bbc.com OR site:cnn.com OR site:news24.com"
url = f"https://api.exa.ai/v1/search?q={query}&api_key={api_key}"
try:
response = requests.get(url)
data = response.json()
if "results" in data and len(data["results"]) > 0:
top_result = data["results"][0] # Get the first search result
title = top_result["title"]
snippet = top_result["snippet"]
link = top_result["url"]
return f"Latest Cape Town weather report:\n{title}\n{snippet}\nMore info: {link}"
else:
return "No weather updates found. Try again later."
except Exception as e:
return f"Error fetching data: {str(e)}"
# Load model from Hugging Face
model = HfApiModel(
model_id='mistralai/Mistral-7B-Instruct', # Change to your preferred model
max_tokens=200,
temperature=0.5
)
# Load prompt templates
with open("prompts.yaml", "r") as stream:
prompt_templates = yaml.safe_load(stream)
# Define the AI Agent
agent = CodeAgent(
model=model,
tools=[search_cape_town_weather], # Add the EXA search tool
max_steps=3,
verbosity_level=1,
prompt_templates=prompt_templates
)
# Test the agent (Optional)
if __name__ == "__main__":
print(agent.run("Find the latest weather update for Cape Town."))
|