# CVE-2026-35029: LiteLLM - Information Disclosure

> Live exploitation tracking for CVE-2026-35029 by the CrowdSec Live Exploit Tracker, built from real-world attack data crowdsourced from CrowdSec's global network.

Page: https://tracker.crowdsec.net/cves/CVE-2026-35029

## Key facts

- **Exploitation phase:** Limited Exploitation. The vulnerability is known but shows very limited attacker interest or exploitation activity.
- **CVSS score:** 8.7
- **Public exploit available:** Yes
- **Affected products:** BerriAI LiteLLM
- **Weaknesses:** CWE-863 (Incorrect Authorization)
- **Tags:** ai_mcp, web_application

## Description

LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. Prior to 1.83.0, the /config/update endpoint does not enforce admin role authorization. A user who is already authenticated into the platform can then use this endpoint to modify proxy configuration and environment variables, register custom pass-through endpoint handlers pointing to attacker-controlled Python code, achieving remote code execution, read arbitrary server files by setting UI_LOGO_PATH and fetching via /get_image, and take over other privileged accounts by overwriting UI_USERNAME and UI_PASSWORD environment variables. Fixed in v1.83.0.

## CrowdSec analysis

[CVE-2026-35029](https://euvd.enisa.europa.eu/vulnerability/CVE-2026-35029) is a high-severity vulnerability in BerriAI's LiteLLM proxy server that allows authenticated users to access the /config/update endpoint without admin authorization. Exploiting this flaw, attackers can modify proxy configurations, execute arbitrary Python code, read sensitive server files, and even take over privileged accounts. This vulnerability could lead to remote code execution, information disclosure, and full compromise of the affected system. The issue is resolved in version 1.83.0.

CrowdSec has been tracking this vulnerability and its exploits since 6th of May 2026.

CrowdSec has not observed any significant exploitation activity targeting CVE-2026-35029 across its network. As a result, no community-driven trend analysis is available for this vulnerability at this time.

Attackers exploit the `/config/update` endpoint by sending crafted POST requests to configure the application to read arbitrary files, such as `/etc/passwd`, and exfiltrate their contents via attacker-controlled endpoints. This attack requires authentication but does not enforce admin role checks, allowing privilege escalation and sensitive file disclosure.

## Timeline

- 2026-04-06: CVE Published. CVE-2026-35029 is published to NVD.
- 2026-05-06: Rule Released. CrowdSec releases a rule to detect CVE-2026-35029 exploitation attempts against the CrowdSec Network.
- 2026-06-09: CrowdSec First Seen. CrowdSec observes in-the-wild exploitation of CVE-2026-35029 for the first time.

## References

- https://github.com/BerriAI/LiteLLM/security/advisories/GHSA-53mr-6c8q-9789
- http://seclists.org/fulldisclosure/2026/Apr/17
- https://github.com/projectdiscovery/nuclei-templates/blob/main/http/cves/2026/CVE-2026-35029.yaml

## Live exploitation data

The [Live Exploit Tracker page](https://tracker.crowdsec.net/cves/CVE-2026-35029) follows observed exploitation activity for CVE-2026-35029. [Request an API key](https://tracker.crowdsec.net/request-api-key) for full access to its exploitation timeline, attacking IP addresses and blocklists.