Censys Research

https://censys.com/resources/blog/

Internet-scale scanning telemetry; rapid-response advisories on edge-device CVEs (added 2026-05-08). 2026-05-08 audit: WebFetch returned 5 dated articles Apr 22 - May 6 2026. Rich complement to GreyNoise / Shadowserver. RSS at https://censys.com/feed/ also works. Candidate; promote to active after 3 runs. | 2026-06-20 full audit (v2.62): live=Y, drill=Y. FETCH → rss: python3 tools/fetch_source.py feed https://censys.com/feed/ 5 (then WebFetch the per-article /blog/<slug>/ URL for the body). AVOID: No need to scrape the /resources/blog/ listing; the feed is clean dated XML and WebFetch on article pages renders fine.. | 2026-07-05 admiralty audit: B (HIGH->B), Censys vendor research, first-hand scan telemetry; keep active. Feed mixes research with company announcements; filter for the /blog/ research items. | 2026-08-05: BLOCKED on every transport. censys.com now returns HTTP 403 to the direct browser-UA fetch AND relays an upstream block/challenge through the jina reader, on the feed (https://censys.com/feed/), the blog index (/blog, /resources/blog/) and even /sitemap.xml, so the whole origin is challenge-gated, not one path. The legacy www.censys.io/blog/rss.xml returns 404. The documented feed recipe is therefore dead until the origin relaxes. Set fetch_method=blocked so source_health.py classes this handled rather than re-flagging it every run; status stays ACTIVE because a 403 is a transport block, never a content-death signal. Re-probe the feed each run and restore fetch_method=rss the moment it clears. | 2026-08-15: feed reachable again on the direct transport after being pinned fetch_method: blocked; switched back to webfetch, re-pin only if it regresses. | 2026-10-03 S1: `feed https://censys.com/feed/ 8` works on the direct transport with dated items; advisory pages (https://censys.com/advisory/<slug>/) read with `extract`. (2026-10-03T0404Z-intel)