View Ridge Security
Back to Cyber Hose

Escape found stored XSS in two AI chatboxes via Markdown rendering

Stored XSS in two AI chat products headlines this run, plus Peer2Profit exposing internal networks and a MISP parser confusion CVE.

🕵️ RESEARCH & DEEP DIVES

  • Escape found stored XSS in two AI chatboxes via Markdown rendering
    Escape found stored XSS in two customer-facing AI chatboxes.

    • The issue affected two customer-facing AI chat products from unrelated companies.
    • Chat messages containing model-generated Markdown could execute JavaScript in users’ browsers.
    • The attack used raw HTML rendering without a sanitizer in the Markdown frontend.
    • Escape’s AI pentesting agent induced the chat models to emit the malicious content.
    Sources 📎 Coverage: securityboulevard.com · 👁 via securityboulevard.com (discovered)
  • Peer2Profit Turns Employee Devices Into Gateways to Internal Networks
    Peer2Profit can expose internal network resources through residential proxy networks.

    • Corporate and remote-work devices running Peer2Profit are affected.
    • Peer2Profit supplies user bandwidth to AstroProxy’s residential, mobile, and datacenter proxy pools.
    • Silent Push identified 117,224 unique AstroProxy IPs over 72 hours, including 60,247 residential IPs.
    • The app is installed with user consent through official channels and may evade antivirus detection.
    • AstroProxy’s private-IP filter was bypassed with a domain resolving to an internal IP address, exposing a router interface through an enrolled proxy node.
    Sources 📄 Source: silentpush.com · 📎 Coverage: cyberpress.org · 👁 via Cyber Security News

🔓 CVEs & KEV

  • CVE-2026-77710 — CVSS 6.9 — STIX2 Parser Confusion and Mass Assignment Allow Unauthorized MISP Attribute ...

Need help assessing your exposure?

Start with the free Posture Self-Check to see where you stand against the current threat landscape.

Free Posture Self-Check