LiteLLM has a sandbox escape in custom-code guardrail
CVE-2026-40217 · GHSA-wxxx-gvqv-xp7p · published 2026-05-11
- Severity
- High
- CVSS score
- 8.8
- Chance of exploitation (EPSS)
- 17.0%
- Fixed in
- 1.83.10
Known exploited
CISA added this to its Known Exploited Vulnerabilities catalog on 2026-11-18; US federal agencies have until 2026-12-02 to act. CISA's required action: Apply mitigations in accordance with vendor instructions, ensuring compliance with CISA’s BOD 26-04 Prioritizing Security Updates Based on Risk.
- Exploitation Public proof of concept
- Automatable No
- Technical impact Total
- CISA SSVC
The POST /guardrails/testcustomcode endpoint runs user-supplied Python inside a hand-rolled sandbox. The sandbox can be escaped using bytecode-level techniques, allowing arbitrary code execution in the proxy process — which runs as root in the default Docker image. Reaching the endpoint requires a proxy-admin credential in default configurations.
Upgrade litellm to 1.83.10 or later. If you start it from an MCP config with npx or uvx, pin that version so an old cached copy is not used.
Fixed in 1.83.11. The hand-rolled sandbox has been replaced with RestrictedPython. Upgrade to 1.83.11 or later.
If upgrading is not immediately possible, block POST /guardrails/testcustomcode at your reverse proxy or API gateway.
| Package | Registry | Affected | Fixed in |
|---|---|---|---|
| litellm | PyPI | >= 1.81.8, < 1.83.10 | 1.83.10 |
- Criticaln8n Vulnerable to Remote Code Execution via Expression Injectionn8n, n8nExploited
- HighLangflow affected by Remote Code Execution via validate_code() exec()langflowExploited
- CriticalJupyter MCP Server: Browser DNS rebinding can hijack the Jupyter backend through management routesjupyter-mcp-server
- Criticalmem0ai Databricks vector store vulnerable to SQL injectionmem0ai
- CriticalFlowise NodeVM sandbox escape via puppeteer allowlist - authenticated RCE and arbitrary file read via Chromiumflowise, flowise-components
- CriticalLangflow: Weak Fernet Key via random.seed()langflow
- CriticalLiteLLM has SQL Injection in Proxy API key verificationlitellmExploited
- HighLiteLLM: MCP Authentication Bypass via OAuth2 Passthrough FallbacklitellmExploited
- HighLiteLLM: Authenticated command execution via MCP stdio test endpointslitellmExploited
- CriticalLiteLLM vulnerable to server-side template injection in the /prompts/test endpointlitellm
- CriticalMalicious code in litellm (PyPI)litellm, litellm, litellm
- CriticalLiteLLM: Authentication Bypass via Host Header Injectionlitellm
The impact, patch and workaround text comes from the advisory, shortened (GitHub Advisory Database, CC BY 4.0). Proof-of-concept and reproduction sections and code are left out; the original advisory has them.