I tried to make a “collect all website links agent” using a scraper + python script, but I get this error:
I let the “Generate” button create this script for me:
import re
from typing import List, Tuple
def extract_urls(text: str) -> List[str]:
"""Extract URLs from text using regex pattern matching."""
if not isinstance(text, str):
text = str(text)
# Use regex to find all URLs
pattern = r'(https?://\S+|www\.\S+)'
urls = re.findall(pattern, text)
# Clean URLs by removing trailing punctuation
cleaned_urls = []
for url in urls:
# Remove trailing punctuation
url = re.sub(r'[.,!?]+$', '', url)
cleaned_urls.append(url)
return cleaned_urls
def extract_and_process_urls() -> None:
"""Main function to extract and process URLs from text."""
# Get input text from specified variable
input_text = ai.getConfig("inputVariable")
if not input_text:
print("No input text provided")
# Set empty outputs
ai.vars[ai.getConfig("urlsOutputVariable")] = ""
ai.vars[ai.getConfig("countOutputVariable")] = 0
return
# Extract URLs
urls = extract_urls(input_text)
# Remove duplicates while preserving order
seen = set()
unique_urls = []
for url in urls:
if url not in seen:
unique_urls.append(url)
seen.add(url)
# Create CSV string
csv_output = "\n".join(unique_urls)
# Store outputs in specified variables
ai.vars[ai.getConfig("urlsOutputVariable")] = csv_output
ai.vars[ai.getConfig("countOutputVariable")] = len(unique_urls)
print(f"Found {len(unique_urls)} unique URLs")
# Execute the function
extract_and_process_urls()
I tried creating different scripts with python and javascript but they all ran into a similar wierd error. Each of them mentioned a “slice”, whatever that means. 