Copyright © 2012 Red Hat, Inc. All rights reserved. Important 2026-07-08T19:32:11 litellm: LiteLLM: Directory traversal via crafted skill archive upload 8.1 CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:H CWE-22

LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. Prior to 1.83.7-stable, LiteLLM Skills archive extraction did not sufficiently validate file paths from uploaded skill ZIP archives, allowing an authenticated user with access to LiteLLM LLM API routes or a key whose allowed_routes includes /v1/skills, anthropic_routes, or llm_api_routes to upload a crafted skill archive containing path traversal entries that could be written outside the intended extraction or staging directory. This issue is fixed in version 1.83.7-stable.

A flaw was found in LiteLLM, a proxy server for Large Language Model (LLM) APIs. An authenticated user, with specific API route access, could upload a specially crafted skill archive. This archive, containing path traversal entries, would allow files to be written outside of the designated extraction directory. This vulnerability could lead to arbitrary file creation or modification, potentially resulting in system compromise.

LiteLLM prior to 1.83.7-stable does not sufficiently validate file paths extracted from uploaded Skills ZIP archives, allowing path traversal writes outside the intended extraction directory. Exploitation requires the LiteLLM Proxy Server's Skills API (/v1/skills, anthropic_routes, or llm_api_routes) to be enabled and reachable by an authenticated caller. Fixed in 1.83.7-stable. Upgrade to LiteLLM 1.83.7-stable or later. Products that use LiteLLM only as a Python SDK (litellm.completion()) rather than running the LiteLLM Proxy Server with the Skills feature enabled are not exposed to this vulnerability, as the vulnerable HTTP upload endpoint does not exist in SDK-only usage. Exploit Intelligence Affected exploit-intelligence-tech-preview/vulnerability-analysis-rhel9 Lightspeed Core Not affected lightspeed-core/lightspeed-stack-rhel9 Red Hat Ansible Automation Platform 2 Not affected ansible-automation-platform-26/lightspeed-chatbot-rhel9 Red Hat Ansible Automation Platform 2 Not affected ansible-automation-platform-27/lightspeed-chatbot-rhel9 Red Hat OpenShift AI (RHOAI) Not affected rhoai/odh-llama-stack-core-rhel9 Red Hat OpenShift AI (RHOAI) Not affected rhoai/odh-mlflow-rhel9 Red Hat OpenShift AI (RHOAI) Not affected rhoai/odh-trustyai-garak-lls-provider-dsp-rhel9 https://www.cve.org/CVERecord?id=CVE-2026-59820 https://nvd.nist.gov/vuln/detail/CVE-2026-59820 https://github.com/BerriAI/litellm/commit/6a15adcd64137d37f73dee76dfe7481f8c2d9196 https://github.com/BerriAI/litellm/pull/25475 https://github.com/BerriAI/litellm/releases/tag/v1.83.7-stable https://github.com/BerriAI/litellm/security/advisories/GHSA-5jmr-gcrj-2c9q