sf-docs

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and scrapes public, third‑party Salesforce sites (e.g., developer.salesforce.com, help.salesforce.com, resources.docs.salesforce.com) via components like scripts/salesforce_dom_scraper.mjs, sync_sf_docs.py, and retrieve_sf_docs.py (run_browser_scraper/fetch_url/live‑scrape flows) and then reads and uses that content to drive retrieval, evidence scoring, and follow‑up actions, so untrusted page content can materially influence agent behavior.