Use their Atlassian Python API to set labels for the children of a given page. The script asks for the URL, and the list of tags. It then finds all children for that page, regardless of URL form, and sets the proper tags.

add_labels_to_page_children.py

import sys
import os
from atlassian import Confluence

def obtain_confluence_pat(path='~/.ConfluencePAT'):
    confluence_PAT_filename = os.path.expanduser(path)
    confluence_PAT = None
    if os.path.exists(confluence_PAT_filename):
        with open(confluence_PAT_filename, "r") as pat_file:
            confluence_PAT = pat_file.readline()
    else:
        from getpass import getpass
        confluence_PAT = getpass(prompt="Confluence Personal Access Token")

    return confluence_PAT

def process_tags(page_id, cf_instance=None, tags=[]):
    for tag in tags:
        cf_instance.set_page_label(page_id, tag)

def add_tags_to_page_children(page_id, cf_instance=None, tags=[]):
    page_children = cf_instance.get_child_pages(page_id=page_id)
    for children in page_children:
        process_tags(children['id'], cf_instance=cf_instance, tags=tags)

def get_id_for_url(url="", cf_instance=None):
    url_items = url.split("/")
    try:
        space_start = url_items.index("display") + 1
        space = url_items[space_start]
        title = url_items[-1]
        page_id = cf_instance.get_page_id(space, unquote_plus(title))
    except ValueError:
        page_id = url_items[-1].split("=")[-1]
    return page_id


if __name__ == '__main__':
    alma_cf_url = 'https://confluence.alma.cl/'
    alma_cf = Confluence(alma_cf_url, token = obtain_confluence_pat())
    
    page_url = input("Provide page URL: ")
    page_id  = get_id_for_url(page_url)
    
    tags = []
    while True:
        tag = input("Tag to add (empty to stop): ")
        if tag != "":
            tags.append(tag)
        else:
            break
        
    add_tags_to_page_children(page_id, cf_instance=alma_cf, tags=tags)