Sign in
Deutsch English Español Polski
High

LiteLLM: Authenticated command execution via MCP stdio test endpoints

  • Disclosed: 25.04.2026
  • Updated: 11.10.2026 15:17

CISA confirms exploitation of this vulnerability in real attacks. Prioritise checking whether your application is affected. CISA KEV · CC0 1.0

Impact

Two endpoints used to preview an MCP server before saving it — POST /mcp-rest/test/connection and POST /mcp-rest/test/tools/list — accepted a full server configuration in the request body, including the command, args, and env fields used by the stdio transport. When called with a stdio configuration, the endpoints attempted to connect, which spawned the supplied command as a subprocess on the proxy host with the privileges of the proxy process.

The endpoints were gated only by a valid proxy API key, with no role check. Any authenticated user — including holders of low-privilege internal-user keys — could therefore run arbitrary commands on the host.

Patches

Fixed in 1.83.7. Both test endpoints now require the PROXY_ADMIN role, bringing them into line with the save endpoint.

Workarounds

If upgrading is not immediately possible, developers should block POST /mcp-rest/test/connection and POST /mcp-rest/test/tools/list at their reverse proxy or API gateway.

Affected software

Click a technology to see all vulnerabilities affecting it.

PlatformPackageVulnerable versionsFixed in
Python (PyPI) litellm >= 1.74.2, < 1.83.7 1.83.7

References

CVSS score 8.8/10
CVSS vector
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Weakness (CWE)
CWE-77, CWE-78
Platform
Python (PyPI)
Package
litellm
Disclosed
25.04.2026
Updated
11.10.2026 15:17

Sources and licenses

OSV / GitHub Advisory Database - PYSEC-2026-2599 - PyPA Advisory Database; CC BY 4.0

OSV / GitHub Advisory Database - GHSA-v4p8-mg3p-g94g - CC BY 4.0

Vulnerability history: litellm

All published vulnerabilities of this package (Python (PyPI)), newest first. Shows whether this is a one-off slip or a package that needs watching.

All (42) in the archive
Total entries
42
First → latest
04.2024 09.2026
Without a fix
14
Exploited in attacks
3

Severity breakdown

  • Critical 7
  • High 17
  • Medium 8
  • Low 9
  • Unrated 1

Disclosures per year

Severity:
  1. LiteLLM vulnerable to server-side template injection in the /prompts/test endpoint
    Critical · 9.8 CVE-2026-37004 < 1.83.7 1.83.7
  2. LiteLLM: MCP Authentication Bypass via OAuth2 Passthrough Fallback
    High Exploited in attacks CVE-2026-59822 < 1.84.0 1.84.0
  3. LiteLLM: Authentication Bypass via Host Header Injection
    Critical · 9.8 CVE-2026-49468 < 1.84.0 1.84.0
  4. LiteLLM allows a user to modify their own user_role via the /user/update endpoint
    High · 8.8 CVE-2026-47102 < 1.83.10 1.83.10
  5. LiteLLM allows an authenticated internal_user to create API keys with access to routes that their role does not permit
    High · 8.8 CVE-2026-47101 < 1.83.14 1.83.14
  6. LiteLLM has a sandbox escape in custom-code guardrail
    High · 8.8 CVE-2026-40217 >= 1.81.8, < 1.83.10 1.83.10
  7. LiteLLM: Authenticated command execution via MCP stdio test endpoints this entry
    High · 8.8 Exploited in attacks CVE-2026-42271 >= 1.74.2, < 1.83.7 1.83.7
  8. LiteLLM has SQL Injection in Proxy API key verification
    Critical · 9.8 Exploited in attacks CVE-2026-42208 >= 1.81.16, < 1.83.7 1.83.7
  9. LiteLLM: Server-Side Template Injection in /prompts/test endpoint
    High CVE-2026-42203 >= 1.80.5, < 1.83.7 1.83.7
  10. LiteLLM: Password hash exposure and pass-the-hash authentication bypass
    High ghsa-69x8-hrgq-fjj8 < 1.83.0 1.83.0
  11. LiteLLM: Authentication bypass via OIDC userinfo cache key collision
    Critical CVE-2026-35030 < 1.83.0 1.83.0
  12. LiteLLM: Privilege escalation via unrestricted proxy configuration endpoint
    High CVE-2026-35029 < 1.83.0 1.83.0
  13. Two LiteLLM versions published containing credential harvesting malware
    Critical ghsa-5mg7-485q-xm76 >= 1.82.7, <= 1.82.8 no fix
  14. LiteLLM Has an Improper Authorization Vulnerability
    High · 8.1 CVE-2025-0628 < 1.61.15 1.61.15
  15. LiteLLM Has a Leakage of Langfuse API Keys
    High · 7.5 CVE-2025-0330 <= 1.52.1 no fix
  16. LiteLLM Reveals Portion of API Key via a Logging File
    High · 7.5 CVE-2024-9606 < 1.44.12 1.44.12
  17. LiteLLM Vulnerable to Denial of Service (DoS) via Crafted HTTP Request
    High · 7.5 CVE-2024-8984 < 1.56.2 1.56.2
  18. LiteLLM Vulnerable to Remote Code Execution (RCE)
    High · 8.8 CVE-2024-6825 >= 1.40.3.dev2, <= 1.40.12 no fix
  19. LiteLLM Vulnerable to Denial of Service (DoS)
    High · 7.5 CVE-2024-10188 < 1.53.1.dev1 1.53.1.dev1
  20. LiteLLM Server-Side Request Forgery (SSRF) vulnerability
    High · 7.5 CVE-2024-6587 < 1.44.8 1.44.8
  21. litellm vulnerable to remote code execution based on using eval unsafely
    Critical · 9.8 CVE-2024-5751 < 1.40.16 1.40.16
  22. Arbitrary file deletion in litellm
    High · 6.5 CVE-2024-4888 < 1.35.36 1.35.36
  23. litellm passes untrusted data to `eval` function without sanitization
    High · 7.2 CVE-2024-4264 <= 1.28.11 no fix
  24. LiteLLM has Server-Side Template Injection vulnerability in /completions endpoint
    Critical · 9.8 CVE-2024-2952 < 1.34.42 1.34.42

Know which vulnerabilities affect your application

With the DockRay integration and software inventory enabled, DockRay checks reported component versions against known vulnerabilities and notifies you about confirmed matches. Coverage depends on the supported technology and available advisory data.

Connect your application
DockRay newsletter

Application security in your inbox

Important vulnerabilities, actionable guidance and practical advice for your applications. Choose your technologies and follow what matters to your project.

  • A daily digest of new vulnerabilities, critical ones included.
  • Notifications about important updates.
  • Practical advice on application security and monitoring.

Leave empty to receive updates for every technology.

Privacy policy

Free to subscribe. Unsubscribe with one click in any email.