CVE-2024-2952

moderate-risk
Published 2024-04-10

BerriAI/litellm is vulnerable to Server-Side Template Injection (SSTI) via the `/completions` endpoint. The vulnerability arises from the `hf_chat_template` method processing the `chat_template` parameter from the `tokenizer_config.json` file through the Jinja template engine without proper sanitization. Attackers can exploit this by crafting malicious `tokenizer_config.json` files that execute arbitrary code on the server.

Do I need to act?

~
1.4% chance of exploitation in next 30 days
EPSS score — moderate exploit probability
-
Not on CISA KEV list
No confirmed active exploitation reported to CISA
?
Patch status unknown
Check vendor advisories for fix availability and mitigation guidance
9
CVSS 9.8/10 Critical
NETWORK / LOW complexity

Affected Products (1)

Litellm

Affected Vendors

41
/ 100
moderate-risk
Severity 32/34 · Critical
Exploitability 4/34 · Minimal
Exposure 5/34 · Minimal