1,852 exposed AI pentest consoles found on Shodan: ARTEX and 20+ renamed variants fingerprintable via Censys and FOFA
cyb3rops · x · 2026-11-05
Security researchers @SansLimit3 and @rstcloud show that ARTEX—the AI harness behind recent intrusions—and its many renamed forks are widely exposed to the internet and easily fingerprinted:
- Shodan: searching title:"自主渗透测试控制台" (plus Korean variants) returns 1,852 instances — ARTEX itself accounts for 1,708, AntiFox 94
- Censys: 1,653 hosts matchable via http.htmltitle regex
- FOFA: 2,000+ results
- Notable variants include AntiFox (typically on port 443 with antifox..nip.io / .sslip.io certs), Autonomous Pentest Agent, LMTex, SHRIKE, SKARN, and 20+ more
- The quoted post also shares hunting syntax for Cairn (title + uvicorn) and PentAGI (favicon hash)
Key caveat: title-only searches miss forks and renames, so a multi-signal approach is required. The scale of exposed consoles also underscores how poorly secured these autonomous pentest deployments are.
More from Safety
- If an agent misuses my valid credentials, what can tell it wasn't me? — Thessaly-Sorties · 2026-11-05
- After Binance's $6.3bn Hack, Slop Cash Tries Token-Funded Open-Source Maintenance — 0xSammy · 2026-11-05
- Emergent misalignment published in Nature: narrow finetuning causes broad LLM misalignment — davidmanheim · 2026-11-05
- Uber open-sources ADR, its production agent security system with 300-task benchmark — thisguyknowsai · 2026-11-05
- Fintech analyst warns banks need a Plan B for AI sovereignty kill-switch risk — mikeflache · 2026-11-05
- Floridi et al. propose portable spec for agent harnesses, split into 8 components — mikeflache · 2026-11-05