An authenticated attacker with administrative access to the appliance can inject malicious JavaScript code inside the definition of a Threat Intelligence rule, that will later be executed by another legitimate user viewing the details of such a rule.
An attacker may be able to perform unauthorized actions on behalf of legitimate users. JavaScript injection was possible in the content for Yara rules, while limited HTML injection has been proven for packet and STYX rules.
The injected code will be executed in the context of the authenticated victim's session.

References

• https://nvd.nist.gov/vuln/detail/CVE-2023-22843
• https://security.nozominetworks.com/NN-2023:4-01