CVE-2024-5826: critical vulnerability in vanna-ai/vanna
Remote Code Execution via Prompt Injection in vanna-ai/vanna
Published · Updated
E-mail alert when it lands in KEV, gets a public PoC or its score goes up.
28Vexday Risk Score
No sign of exploitation. No public exploitation artifact known so far.
ssvc Trackcvss 9.8epss 0.9%
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.
CVSS:3.0/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Affected products
vanna-ai · vanna-ai/vannaRelated CVEs — vanna-ai/vanna
In the same product, most dangerous first.
CVE-2024-5827CRITICALArbitrary File Write by Prompt Injection via DuckDB SQL in vanna-ai/vannaEPSS 3.4%CVE-2024-7764HIGHSQL Injection in vanna-ai/vannaEPSS 0.8%CVE-2024-8055HIGHLocal File Read (LFI) by Prompt Injection via SnowFlake SQL in vanna-ai/vannaEPSS 0.7%CVE-2024-5753HIGHLocal File Read (LFI) by Prompt Injection via Postgres SQL in vanna-ai/vannaEPSS 0.6%CVE-2024-8099HIGHServer-Side Request Forgery (SSRF) in vanna-ai/vannaEPSS 0.4%CVE-2024-6841MEDIUMCSRF in vanna-ai/vannaEPSS 0.2%