PT-2026-53631 · Pypi · Vanna

Published 2026-06-29 · Updated 2026-06-29

OSV

CVSS v3.1

9.8

Critical

Base vector

AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

In the latest version of vanna-ai/vanna, the vanna.ask function is vulnerable to remote code execution due to prompt injection. The root cause is the lack of a sandbox when executing LLM-generated code, allowing an attacker to manipulate the code executed by the exec function in src/vanna/base/base.py. This vulnerability can be exploited by an attacker to achieve remote code execution on the app backend server, potentially gaining full control of the server.

Fix

Related Identifiers

PYSEC-2026-562

Affected Products

Vanna

References · 6

• https://osv.dev/vulnerability/PYSEC-2026-562 · Vendor Advisory
• https://nvd.nist.gov/vuln/detail/CVE-2024-5826 · Security Note
• https://github.com/vanna-ai/vanna · Note
• https://pypi.org/project/vanna · Note
• https://huntr.com/bounties/90620087-44ac-4e43-b659-3c5d30889369 · Note
• https://github.com/advisories/GHSA-rrqq-fv6m-692m · Note