CVE-2026-12797: BerriAI litellm Completions banned_keywords.py async_pre_call_hook authorization

Published Jun 21, 2026
·
Updated

A security flaw has been discovered in BerriAI litellm up to 1.82.5. Affected is the function asyncprecallhook of the file enterprise/enterprisehooks/bannedkeywords.py of the component Completions Interface. The manipulation of the argument prompt results in incorrect authorization. The attack may be performed from remote. The exploit has been released to the public and may be used for attacks. The vendor was contacted early about this disclosure.

Affected Software

2 affected components
pypi/litellm<=1.82.5
LiteLLM LiteLLM<=1.82.5

Remediation

Recommended actions to resolve this vulnerability, in priority order.

  1. Compensating control

    Restrict network access to the Completions Interface (enterprise/enterprise_hooks/banned_keywords.py async_pre_call_hook) to trusted IPs/networks only using firewall rules, VPC security groups, or WAF access controls to prevent remote exploitation.

  2. Compensating control

    Deploy input-validation/WAF rules on the Completions Interface to inspect and reject requests that manipulate the 'prompt' argument (block prompt values or payloads that deviate from expected format or include unauthorized modifications) to mitigate the incorrect-authorization vulnerability in async_pre_call_hook.

  3. Operational

    Audit and monitor logs for calls to async_pre_call_hook in enterprise/enterprise_hooks/banned_keywords.py; investigate and block suspicious remote requests that manipulate the 'prompt' argument. If unauthorized access is suspected, block offending IPs and revoke/rotate any exposed API keys or credentials associated with the Completions Interface.

Event History

Jun 21, 2026
CVE Published
via MITRE·09:15 AM
Data Sourced
via MITRE·09:15 AM
DescriptionSeverityWeakness
Data Sourced
via NVD·10:16 AM
DescriptionSeverityWeaknessAffected Software
Free Weekly Intel

Don't miss critical vulnerabilities

Join thousands of security professionals who receive our weekly digest of trending CVEs, zero-days, and exploited vulnerabilities.

No spam. Unsubscribe anytime.

Frequently Asked Questions

1

What is the severity of CVE-2026-12797?

The severity of CVE-2026-12797 is classified as medium with a score of 6.3.

2

How do I fix CVE-2026-12797?

To fix CVE-2026-12797, update BerriAI litellm to the latest version that mitigates this vulnerability.

3

What are the potential impacts of CVE-2026-12797?

CVE-2026-12797 may lead to incorrect authorization by manipulating user input in the async_pre_call_hook function.

4

In which component is CVE-2026-12797 found?

CVE-2026-12797 is found in the Completions Interface of the BerriAI litellm under the file enterprise/enterprise_hooks/banned_keywords.py.

5

What is the affected software version for CVE-2026-12797?

CVE-2026-12797 affects BerriAI litellm versions up to 1.82.5.

Contact

SecAlerts Pty Ltd.
132 Wickham Terrace
Fortitude Valley,
QLD 4006, Australia
info@secalerts.co
By using SecAlerts services, you agree to our services end-user license agreement. This website is safeguarded by reCAPTCHA and governed by the Google Privacy Policy and Terms of Service. All names, logos, and brands of products are owned by their respective owners, and any usage of these names, logos, and brands for identification purposes only does not imply endorsement. If you possess any content that requires removal, please get in touch with us.
© 2026 SecAlerts Pty Ltd.
ABN: 70 645 966 203, ACN: 645 966 203