CVE-2026-96893 PUBLISHED

Improper Neutralization of Input Used for LLM Prompting in GitLab

Assigner: GitLab
Reserved: 23.09.2026 Published: 13.10.2026 Updated: 13.10.2026

GitLab has remediated an issue in GitLab EE affecting all versions from 19.1 before 19.2.8, 19.3 before 19.3.5, and 19.4 before 19.4.2 that, under certain conditions, could have allowed an authenticated user to inject specially crafted content into AI agent instructions, potentially causing unintended command execution attributed to other users, due to improper neutralization of special elements in AI conversation context construction.

Metrics

CVSS Vector: CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:U/C:H/I:H/A:N
CVSS Score: 7.3

Product Status

Vendor GitLab
Product GitLab
Versions Default: unaffected
  • affected from 19.1 to 19.2.8 (excl.)
  • affected from 19.3 to 19.3.5 (excl.)
  • affected from 19.4 to 19.4.2 (excl.)

Solutions

Upgrade to versions 19.2.8, 19.3.5, 19.4.2 or above.

Credits

  • This vulnerability has been discovered internally by GitLab team member Mert Coskuner. finder

References

Problem Types

  • CWE-1427: Improper Neutralization of Input Used for LLM Prompting CWE