CVE-2026-42271
Named by 7 kept items in this workspace.
Search every workspace for CVE-2026-42271. It matches the name as written, so it can find other things that share it.
The items that name it 7
Command injection in BerriAI LiteLLM Model Context Protocol (MCP) server preview endpoint (POST /mcp-rest/test/connection), resulting in host takeover and API credential theft. Rated High Threat Risk; confirmed exploited in the wild.
Command injection in LiteLLM's MCP server test endpoints; the command field in MCP stdio server test configurations is passed directly to subprocess execution with no validation, enabling remote code execution. Added to CISA KEV in June 2026.
Horizon3.ai public research
authenticated user can run commands on the LiteLLM host; added to CISA's KEV catalog, indicating exploitation in the wild
Command injection vulnerability in BerriAI LiteLLM allowing any authenticated user to run arbitrary commands on the host; CVSS score 8.7. Added to CISA KEV catalog. Affects LiteLLM Python package >= 1.74.2, < 1.83.7. Root cause: POST /mcp-rest/test/connection and POST /mcp-rest/test/tools/list endpo
- CVE-2026-42271: LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format
LiteLLM proxy server vulnerability where POST /mcp-rest/test/connection and POST /mcp-rest/test/tools/list endpoints, used to preview an MCP server before saving it, accepted a full server configuration including command, args, and env fields for stdio transport. Calling them with a stdio configurat
All cves in this workspace.