Web-verified equity research: CNBC/Morningstar/Forbes 403 on plain fetch — use a headless scraper, cross-check aggregator YTD tables against a dated primary

When compiling sourced stock-performance research (episodes + current YTD/12-month numbers) via an agent, the plain URL fetcher returns HTTP 403 on CNBC, Morningstar, Forbes, Axios and Yahoo Finance sub-pages, and aggregator "YTD tables" silently disagree with each other (one showed a 1-year return of +898% for a $5T stock — a data error) and with dated primary articles.

1) Route 403 sites (CNBC, Morningstar, Axios, Forbes) through a headless scraper tool (firecrawl-style) with onlyMainContent — it returns full markdown plus article:published_time metadata, which you need for "as of" dates. 2) Never take a single aggregator's YTD/1y table: pair it with a dated primary (the outlet's own article stating "up X% in 2026 as of <date>") and record both with as-of dates; flag any value that fails a sanity check (1y return vs market cap) as null with a note rather than passing it on. 3) For per-year Mag-7-style return rankings, financial-news roundups published in the first weeks of the following year list every name with the index comparison; year-end infographics and Statista charts often omit half the names. 4) Fire searches in parallel batches of 10-15 (one batch per company-set) and only then fetch the 5-8 pages that carry the numbers; that kept a 10-company, 30-episode research job at ~6 tool rounds.

Agent asked for web-verified, URL-backed research on stock episodes and current performance across many tickers.