CVE-2026-73967
8.7HighPublished 18 Nov 2026 · NVD status: Received · AI triage by gemini-3.1-flash-lite
High
0CVSS v3.110
- Attack vector
- Network
- Complexity
- Low
- Privileges
- Low
- User action
- Required
- Impact
- High impact on confidentiality, integrity
- Weakness
- CWE-79, CWE-922, CWE-1021
What an attacker gets
Authenticated attackers can execute arbitrary scripts in other users' sessions to steal JWT tokens and hijack accounts. This occurs when a user views job output or forms containing malicious payloads injected by an attacker.
Root cause
The application renders user-controlled input using v-html without sanitization and lacks a Content Security Policy to restrict script execution.
Affected versions
AnsibleForms prior to 6.2.1
CVSS v3.1 · CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:N
Fix & remediation
Upgrade to version 6.2.1 or later. Until you can patch, restrict access to the AnsibleForms web interface to trusted users only.
AI-suggested. Run it in staging first and check it against the vendor advisory. Why?
NVD description
AnsibleForms is a self-hosted web application that turns automation into self-service forms. Prior to 6.2.1, AnsibleForms stores attacker-influenceable job output that server/src/lib/common.js processes through formatOutput without HTML escaping and client/src/components/AppAnsibleOutput.vue renders through v-html. An authenticated user can submit job input that is echoed into persistent output, causing script execution when another user views the job. Additional designer-controlled HTML and expression values reach v-html in client/src/components/AppForm.vue and client/src/components/BsInputForForm.vue without sanitization. Because Content Security Policy is disabled and the JWT is stored in localStorage, script execution can steal a victim's token and take over the account, including an administrator account. This issue is fixed in version 6.2.1.