Ansible Automation Platform EUVD-2026-85571

Exploitation requires: (1) a low-privileged authenticated automation-controller account able to produce job, ad hoc command, project update, or inventory update standard output - or the ability to inject data that a playbook echoes into that output; (2) the target output rendered through the HTML stdout view, which escapes HTML metacharacters but does NOT strip ANSI escape sequences, so an OSC 8 hyperlink becomes an unfiltered HTML anchor; (3) the text/html response served with no Content-Security-Policy, permitting a javascript: href to execute; and (4) a higher-privileged user (e.g., admin) both viewing that output page and clicking the crafted link (UI:R). The requirement for a privileged victim to actively click the link is the principal factor limiting real-world exploitation.

This is a genuine priority-tier finding, not a high-CVSS-but-low-risk artifact. The CVSS 3.1 vector (AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:N, base 8.7) describes a network-reachable, low-complexity stored XSS requiring only a low-privileged account and a victim click, with a scope change (S:C) reflecting that attacker JavaScript executes in a higher-privileged user's authenticated session - the description explicitly ends in 'full platform takeover,' justifying C:H/I:H. The absence of any Content-Security-Policy on the text/html output page removes the last mitigating control, making the javascript: payload reliably executable. No EPSS score, no CISA KEV listing, and no public proof-of-concept were provided, so there is no evidence of active or automated exploitation - the severity rests on the mechanism and impact rather than observed in-the-wild activity. The signals are internally consistent: network vector matches a web UI, and PR:L correctly reflects that the attacker must be able to produce job/inventory output (or feed data a playbook echoes). The main gating factor is UI:R - a privileged operator must actually view and click the crafted link - which tempers the real-world likelihood relative to a no-interaction flaw.

A low-privileged automation-controller user runs a job (or supplies data that a playbook echoes into standard output) containing a crafted ANSI OSC 8 hyperlink whose href is a javascript: URI. When an administrator later opens that job's HTML stdout view and clicks the rendered anchor, the attacker's JavaScript runs in the admin's authenticated session, enabling actions up to full platform takeover. No public proof-of-concept was identified at the time of analysis, but the technique is straightforward to reproduce.