bobseares-2/seleniumbase-website-scraper
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

write agent.py

Browse Source
This commit is contained in:
a2a-cloud
2026-06-08 22:49:16 +00:00
parent 21d733b9d1
commit 0b79431716
1 changed files with 7 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
agent.py
View File

@@ -291,6 +291,13 @@ class SeleniumbaseWebsiteScraper(A2AAgent[SeleniumbaseWebsiteScraperConfig, NoAu
same_origin_safeguard = _same_origin same_origin_safeguard = _same_origin
await ctx.emit_progress("starting deep-agent iterative scrape planner") await ctx.emit_progress("starting deep-agent iterative scrape planner")
# Domain guard
def _allowed_origin(url: str) -> bool:
if not allowed_domains:
return same_origin_safeguard(start_url, url)
host = urllib.parse.urlparse(url).netloc.lower()
return any(host.endswith(dom.lower()) for dom in allowed_domains)
# Planner system prompt with strict safety policy # Planner system prompt with strict safety policy
system_prompt = ( system_prompt = (
"You are a cautious scraping planner for a compliant browser scraper. " "You are a cautious scraping planner for a compliant browser scraper. "